imprinting hatchery reared salmon and steelhead...

IMPRINTING HATCHERY REARED SALMON AND STEELHEAD TROUT

FOR HOMING, 1978 - 1983

VOLUME I of III: NARRATIVE

Final Report of Research

Prepared by

Emil SlatickLyle G. GilbreathJerrel R. HarmonTed C. BjornnR.R. Ringe

Kenneth A. WalchAnthony J. Novotny

andWaldo S. Zaugg

Coastal Zone and Estuarine StudiesNorthwest and Alaska Fisheries Center

National Oceanic and Atmospheric Administration2725 Montlake Boulevard EastSeattle, Washington 98112

Funded by

Tom Vogel, Project ManagerU.S. Department of Energy

Bonneville Power AdministrationDivision of Fish and Wildlife

P.O. Box 3621Portland, Oregon 97232

Project No. 78-lContract No. DE-AI79-84BP39646

February 1988

iii

PREFACE

Because of the scope of this report, it was prepared in three separate

volumes. The Narrative is contained in Volume I, Volume II summarizes the

data in tabular form, and Volume III contains the supplemental information on

disease and physiology relating to the juvenile salmonids used in the study.

iv

EXECUTIVE SUMMARY

The National Marine Fisheries Service (NMFS), under contract to the

Bonneville Power Administration, began conducting homing research on Pacific

salmon and steelhead in 1978. The juvenile marking phase of the study was

completed in 1980, and adult returns were examined through 1985. Over 4 million

juvenile salmon and steelhead were marked and released, and 23 individual

experiments were conducted. The research had the following objectives:

(1) develop the techniques for imprinting homing cues while increasing survival

of hatchery reared salmonids and (2) provide fishery managers with the

information necessary to increase the returns of salmon and steelhead to the

Columbia River system and to effectively distribute these fish to the various

user groups.l/

Cur imprint methods were grouped into three general categories:

(1) natural migration imprint from a hatchery of origin or an alternate homing

site (by allowing fish to volitionally travel downstream through the river on

their seaward journey), (2) single exposure imprinting (cueing fish to a single

unique water supply with or without mechanical stimuli prior to transport and

release), and (3) sequential exposure imprinting (cueing fish to two or more

water sources in a step-by-step process to establish a series of signposts for

the route “home”). With variations, all three techniques were used with all

salmonid groups tested: coho salmon, spring and fall chinook salmon, and

steelhead. For the single and sequential imprinting, fish were transported

around a portion of their normal migration route before releasing them into the

Columbia River.

11 Imprinting is defined as a rapid and irreversible learning experience thatprovides fish with the ability to return to natal streams or a preselectedhoming site.

V

There were successes and failures with all the imprint techniques used.

The sequential imprint concept, inherent in both the natural migration imprint

and to a lesser degree the sequential exposure imprint using truck and/or barge

transport, was most successful in imprinting fish. The natural migration

imprint method (used for control releases) was generally effective in imprinting

all stocks of salmon and steelhead used in our experiments. However, this

method was not uniformly successful in returning fish to a homing site.

Differences in homing behavior between steelhead and coho salmon to similar

single imprint methods indicated there were some species differences in response

to homing cues.

Our data indicated that transported indigenous stock or stocks acclimated

to a specific hatchery showed a greater propensity to return to the homing site

than non-indigenous stock fish. This behavior suggests that the genotype of the

fish has a positive influence on the homing cue induced by the imprint

techniques -

The data also indicated that mode of transportation had a significant

impact on the success of imprinting fish for return to a specific homing site.

In general, the comparative survival of fish in the test groups was

enhanced when they were transported around hydroelectric dams on the Snake and

Columbia rivers. In many cases, significantly greater numbers of test than

control fish returned to the Columbia River, as indicated by recoveries at

Bonneville Dam. For example, despite the fact that homing was impaired on a

portion of the transported fish, the return of adult steelhead to the upper

river above McNary Dam (RM 292), in most instances, was as great or greater than

the return of fish from the control releases.

Even though imprint treatments weren’t completely successful in implanting

a homing cue to the entire test population of juveniles being transported

Vi

benefits were possible. The portion of the fish in the test groups that

received a poor or impaired homing cue strayed to other areas- Also, the

portion of fish that were not ready or able to implant a homing cue at the time

they were transported returned as adults to the vicinity of their release site

and remained there. Also in many cases, the enhanced survival resulting from

transportation still provided as many or more fish back to the homing site or

geographic area as returned from the control releases- In general, the net

result of the enhanced survival combined with the modification of adult

migratory behavior induced by the imprint method used provided greater numbers

of fish to the river for sport as well as for native and commercial fisheries

while still providing fish for spawning.

We found that both test and control lots of spring chinook salmon had

extremely high mortality rates whereas fall chinook salmon subjected to the same

handling showed relatively good recovery rates- In the past, Ebel (1973) showed

good survival rates for spring chinook salmon that were marked and transported

from Ice Earbor Dam- It is apparent that a severe problem existed with spring

chinook salmon used in our studies. One possible problem may be bacterial

kidney disease (BKD), common in all the spring chinook salmon stocks used- The

presence of BKD, when combined with the stress induced by handling and marking,

may be the factor that caused the low rates of return.

The effects of imprint strategies on the homing behavior of adult salmonid

migrations plus the enhanced survival produced by transportation provide a tool

that can be used by fisheries managers to provide wre salmon and steelhead to

the various user groups. The homing imprint information on various species and

stocks can be effectively used to rehabilitate fish populations in the tributary

systems of the Columbia River (e.g., the Yakima and Umatilla rivers)- Because

imprinted fish can be induced to return to different river areas, they can be

Vii

manipulated to contribute to specific sport or commercial fisheries that are

presently underdeveloped with existing hatchery releases.

Our results should be viewed as base line information on the reactions of a

given fish stock to the variables within the specific study. Treatments which

have been replicated between different years show similar results. Fishery

researchers and managers can mix, match,, and combine the information on the

results and variables from various treatments contained in this report to

develop additional imprint techniques to achieve desired results. Imprint

strategies will have to be tailored or fine tuned to fit the requirements of the

different species and stocks of salmonids to develop the most effective

techniques for required management goals.

viii

CONTENTS

Page

INTRODUCTION........................................................... 1

Background ........................................................... 1

Objectives ........................................................... 4

STUDY SITES AND FACILITIES ............................................. 4

GENERAL METHODS........................................................ 7

COHO SALMON EXPERIMENTS................................................ 11

Carson NFH - Pasco Homing Site, 1978................................. 13

Background ......................................................... 13

Experimental Design................................................ 13

Results and Discussion ............................................. 14

Conclusion ......................................................... 16

Willard NFH - Stavebolt Creek, 1978.................................. 17

Background ......................................................... 17


Results and Discussion............................................. 21

Conclusions ........................................................ 25

Willard NFH,, 1980.................................................... 27

Background ......................................................... 27



IX

Page

Conclusions ........................................................ 38

Overview of Imprint Treatments of Coho Salmonn ........................ 39

Natural Migration Imprintt .......................................... 39

Single Exposure Imprint ............................................ 39

Sequential Exposure Imprint ........................................ 39

Application of Findings ............................................ 40

SPRING CHINOOK SALMON EXPERIMENTSS ...................................... 41

Kooskia NFH,, 1978.................................................... 41

Background ......................................................... 41


Results ............................................................ 42

Carson NFH, 1979 and 1980............................................ 42

Background......................................................... 42


Results ............................................................ 43

Carson NFH - Pasco, 1979 ............................................. 43

Background......................................................... 43


Results ............................................................ 44

Leavenworth NFH, 1980................................................ 44

Background ......................................................... 44



Kooskia NFH,, 1980.................................................... 49

Background ......................................................... 49

X

Page

Experimental Design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . so

Results ............................................................ 50

Rapid River SFH,, 1980................................................ 51

Background ......................................................... 51


Results ............................................................

Overview of Imprint Treatments of Spring Chinook Salmonn............

FALL CHINOOK SALMON EXPERINENTSS ........................................

Spring Creek NFH - White Salmon - Stavebolt Creek, 1979..............

Background .........................................................

Experimental Design................................................

Results and Discussion.............................................

Conclusions ........................................................

Big Creek SFH - Stavebolt Creek, 1980................................

Background .........................................................

Experimental Design ................................................

Results and Discussion.............................................

Conclusions ........................................................

Spring Creek NFH, 1980...............................................

Background .........................................................

Experimental Design................................................

Results and Discussion .............................................

Conclusions ........................................................

Hagerman NFH, 1980 ...................................................

Background .........................................................

Experimental Design ................................................

52

52

53

53

53

55

56

62

64

64

64

65

70

71

71

72

74

78

79

79

79

xi

Page


Conclusions ........................................................ 82

Overview of Imprint Treatments of Fall Chinook Salmon................ 83

Natural Migration Imprint .......................................... 83

Single Exposure Inprintt ............................................ 84

Sequential Exposure Imprint ........................................ 84


STEELHEAD EXPERIMENTS.................................................. 86

Dworshak NFH,, 1978................................................... 86

Background ......................................................... 86

Experimental Designn................................................ 88

Results and Discussion ............................................. 88

Conclusions ........................................................ 92

Dworshak NFH,, 1980................................................... 93

Background ......................................................... 93



Conclusions ........................................................ 95

Tucannon SFH,, 1978................................................... 97

Background ......................................................... 97



Conclusions .......................................................... 100

Tucannon SFH, 1979 ................................................... 100


xii

Page


Conclusions ........................................................ 104

Tucannon SFH - Little Goose Dam, 1980................................ 106

Background ......................................................... 106


Result and Discussion .............................................. 108

Conclusions ....................................................... 110

Wells SFH-Winthrop NFH, 1978-79 ...................................... 110

Background ......................................................... 110



Conclusions ........................................................ 118

Chelan SFH-Leavenworth NFH, 1978-79 .................................. 119

Background ......................................................... 119



Conclusions ........................................................ 124

Overview of Imprint Treatments of Steelheadd .......................... 125

Natural Migration Imprint .......................................... 125

Single Exposure Imprint ............................................ 130

Sequential Exposure Imprintt ........................................ 132


CONCLUSONS AND RECOMMENDATIONSS ......................................... 135

ACKNOWLEDGMENTSS ........................................................ 137

LITERATURE CITED....................................................... 138

INTRODUCTION

Background

The loss of juvenile migrants in the mainstem Columbia and Snake rivers

is a significant obstacle to the maximum production of salmon and steelhead

trout in the Columbia River Basin. The effects of dams, predation, and

pollution all take an enormous toll (Collins 1976). Losses as high as 95% of

steelhead and chinook salmon were measured from the Salmon River to The Dalles

Dam (Raymond 1979). Losses greater than 50% were measured from the Bonneville

pool to the lower estuary for chinook salmon (Ebel 1970). Similar losses were

indicated for coho salmon. Obviously, a system that would safely bypass fish

around river areas in which these high losses occur would be of immense

benefit to the resource.

A system that collects juvenile migrants at up-river dams and transports

them around the hazards of many dams and impoundments for release in the lower

river iis now operational (Matthews et al. 1977). However, an efficient,

effective, and far more flexible way to protect the hatchery portion of the

downstream migration would be to transport them directly from the hatcheries

without releasing them into the hazards of the river system. Unfortunately ,

previous experiments showed that juvenile fish transported directly from a

hatchery did not “home” back to the hatchery as adults (Taft and Shapovalov

1938 ; Vreeland et al. 1975). Apparently these fish had not received an

adequate homing imprint at the time of their release.

The ability to cue fish to a desired homing location (i.e., hatchery of

origin or any unique water source) has many applications in the management of

salmonids in the Columbia River drainage and other river systems. The ability

to cue juvenile salmonids to a homing location in conjunction with

2

transporting them around hazardous areas during their outmigration should

substantially increase their survival and subsequent return as adults to

hatcheries, streams, and/or harvest areas. Increased returns available to

hatcheries could produce an excess of spawning stock required for egg take.

This "surplus" could be distributed by the fisheries manager by use of

appropriate homing strategies.

The "homing imprint", a rapid and irreversible learning experience that

provides the navigational clues by which salmon and steelhead find their way

back to their natal area, seems to occur at the time smolts are released into

the river system or shortly thereafter (Wagner 1968; Scholz et al. 1975;

Cooper et al. 1976). There are several theories on how the homing imprint

occurs in fish. It is generally accepted, however, that during the freshwater

migration of adult salmonids, odors provide the navigational route to the

home-stream. Studies have indicated that olfactory perception in salmonids is

very acute--fish are able to detect an odor diluted to as little as one part

in 8x10 10 (Alderdice et al. 1954; Idler et al. 1956, 1961; McBride et al.

1962).

Two principal olfactory theories have been proposed. Hordeng (1971,

1977) proposed that the important odors for navigation are population-specific

pheromones from juvenile salmon residing in the river. However, Jensen and

Duncan (1971) demonstrated that adult coho salmon homed to a spring water

source in the Snake River system which did not contain any fish. In returning

to the spring water source, these adults would clearly have had to abandon a

pheromone scented route in the Columbia River which led back to their hatchery

of origin. On the other hand, Hasler and Wisby (1951) proposed that the

environmental (organic and inorganic) odors, which are identifiable over the

3

life span of the species, provide the guidance; this view is supported by

Brannon (1984).

We believe that environmental odors are specific for various tributaries

in the system. This factor, along with unique environmental conditions in

stream systems, probably accounts for the genetic diversity of the numerous

species and racial stock present. Therefore, we chose the environmental odors

concept as most applicable to our experimentation.

Experiments have indicated the possibility of imprinting salmonid smolts

in a hatchery without release into a river system (Madison et al. 1973; Cooper

et al. 1976; Cooper and Scholz 1976). Previous experiments have suggested

that imprinting may be a comparatively rapid phenomenon. In 1971 and 1972

coho and chinook salmon from the Issaquah State Fish Hatchery (SFH) near

Seattle, Washington, were exposed for imprinting purposes to water at the

National Marine Fisheries Service (NMFS) laboratory in Seattle for periods of

time ranging from 4 h to 14 d before release. Adults returned from as little

l/as a 4 h exposure (Groves- ). Studies by Ebel et al. (1973), Ebel (1980)M and

Slatick et al. (1975) have shown that omitting a large portion of a fish's

normal migration route via transportation does not diminish the homing ability

of fish that have smolted and migrated up to several hundred kilometers on

their own volition. It is this area of salmon and steelhead behavior that we

examined in detail.

----------- -----------

11- Al Groves, Biologist, Northwest Region, NMFS, 7600 Sand Point Way N.E.,Seattle, WA 98115, pers. c o m u n . 1986.

4

Objectives

The primary objectives of our homing research were as follows:

1. Determine how to imprint a homing cue in hatchery reared salmonids.

2. Provide fishery managers with information to increase the returns of

salmon and steelhead to the Columbia giver sys ten and to effectively

distribute the fish to the various user groups.

STUDY SITES AND FACILITIES

The homing imprint sites and in-river sampling facilities used were

spread throughout a major portion of the Columbia giver system presently

available to anadromous salmonids (Figs. 1 and 2). Selection of the homing

imprint sites was based on the following criteria: (1) availability of

various stocks of salmon and steelhead at a hatchery which had returning

adults for brood stocks, (2) availability of a hatchery with a unique water

supply which would permit the introduction of other species of salmonids, and

(3) availability of an unmanned site with a unique water supply. All homing

sites were located at permanent facilities (hatcheries) except Stavebolt

Creek, Oregon, and Pasco, Washington, where special facilities were

constructed.

Stavebolt Creek is a tributary to the Lewis and Clark giver which drains

into Youngs Bay near Astoria, Oregon, (Fig. i). The imprinting site was a

pond supplied soley by water from Stavebolt Creek. Pour floating pens 10 x 20

x 6 ft were used to hold the test fish from 4 to 40 h for imprinting. A weir

and trap were constructed near the mouth of Stavebolt Creek to intercept

returning adults.

5

1. Leuwwonh liauhe~ 6Oring chii-SMludl2. Kooskia H~ICIBRIV tScwing ChiwoW3. Rapid Akm Hmhny lswinq cl9imwkl4. lzamln~~chinwkI6. Pare0 c**lr chhmok-cahol6. Scmebdc cm* IFJI Chinook-Cdd7. 9yC.nwkH~ (FHChadl6. Swing Creek IWduw (FaiI Chiil9. 9ig Whir Sdmcn Chmml IF111 Chinook)10. ~orrr Grmite Dvm IFan Chhad11. Wlllud Hmdmw lcdml12. wndww Hmchq ~SmeHNull13. L-r Mahow River I-~14. fvumon Haahwv ~Smelhedl15. Little Goose Dm 6mdhcdl16. Dwonhdl l4ndwrr memndl

Figure 1. --Area map indicating experimental homing sites, 1978-1980.

‘*A7. @a l �r i

i

IIII

I*I

I

I

ZDA l - 5 1 Zone 6I

Commercialgill net fishery

l j<Tmty Indian rrtart firhrryb jI

I II I

Figure 2.--‘Map of Columbia River system showing location of six in-river sampling locations.

7

The Pasco site is located on the first small stream on the east shore of

the Columbia River above the confluence of the Snake River. This stream is

comprised of ground water which surfaces and is pumped from behind levees at

this location. A weir, fish ladder, and three raceways were constructed to

imprint juveniles and recover returning adults.

In-river sampling facilities had traps to intercept tagged adults in the

fishladders at Bonneville, McNary, and Lower Granite dams without having to

sacrifice the fish. The traps consisted of a Denil fishladder leading adults

to a tag detection system which shunted tagged fish into a trap (Ebel 1974;

Slatick 1975) (Fig. 3).

GENERAL METHODS

Juveniles marked in these experiments were from randomized samples or

were selected by the various -hatchery managers to represent the standard fish

produced at their stations. Fish were marked by adipose fin excision and a

magnetic coded wire tag (CWT) (Ebel 1974). Fish (except fall chinook salmon)

expected to return as adults to locations above Bonneville Dam were also

thermal branded (Mighell 1969; Park and Ebel 1974). Branded adults recovered

at in-river traps were identified by brand, jaw tagged to indicate it had been

previously identified, and released to continue their upstream migration (Ebel

et al. 1973; Gilbreath et al. 1976; Slatick 1976). A tank truck [18,900-liter

(5,000-gal) capacity] was used to transport most test fish (Smith and Ebel

1973). A few groups of fish were transported in a barge utilizing a regulated

flow-through water system (McCabe et al. 1979).

Random samples of test fish were taken at the various hatcheries and

maintained in seawater net-pens at the Manchester Marine Experimental Station,

Manchester, Washington. Survival of the sample groups provided a measure of

8

Figure 3.--Plan view and isometric diagrams of wire tag detector andfish separator systems used at Bonneville, McNary. andLower Granite Dams.

9

the level of smoltification (physiological readiness to migrate and adapt to

seawater). A general health survey of the test fish was provided by

additional 60-fish samples taken at the hatcheries and held in fresh water at

the Manchester facility. Blood and tissue samples from these fish were tested

for various disease organisms (Volume III, Novotny and Zaugg 1979, 1981,

1984).

In the normal release of smolts from a hatchery there is a series of

immediate events that may trigger the imprinting of a homing cue. These

events occur at a time when there are major changes in the fish’s physiology

which may make it more sensitive to certain stimuli. The fish are usually

released from a crowded hatchery pool into a completely new environment with

different physical and chemical water qualitites (e.g., temperature, velocity,

turbulence, turbidity, oxygen, odors, etc.); visual phenomena; and spatial

relationships. Any one or more of these changes may initiate imprinting.

Since the phenomenon of elevation in gill sodium, potassium stimulated

ATPase (Na+-K+ ATPase) activity was first reported to be associated with parr-

smolt transformation in steelhead (Zaugg and Wagner 1973) and in coho salmon,

0. kisutch, (Zaugg and McLain 1970), numerous experiments have been conducted

to verify these results and extend observations to other species. As a

result, it has been conclusively shown that the rise in gill Na+-K+ ATPase

activity is one of the many physiological changes which occur at the time of

parr-smolt transformation. We utilized gill Na+-K+ ATPase activity as a

possible biological indicator of the optimum time for imprinting salmonids

with homing cue.

Research was &signed to delineate the period (imprint “window”) when a

juvenile was physiologically ready to accept a long-term homing imprint in its

“memory” and to determine the stimuli or combination of stimuli (defined as an

“event”) required to initiate this memory retention (homing imprint).

10

The techniques used to try and imprint salmon and steelhead are grouped

into three major categories: (i) natural migration imprint, (2) single

exposure imprint, and (3) sequential exposure imprint.

Natural migration imprinting is cueing fish by allowing them to

volitionaliy travel downstream through the river system on their seaward

journey. because of the locations of the homing sites, two distinct release

strategies were employed. In one situation, fish were released directly from

their hatchery of origin or the hatchery in which they were reared. In the

second situation, when the homing site was not at the hatchery of origin, the

fish were either released directly after hauling by truck or held from 4 h to

10 d at the homing site for acclimation and hopefully imprinting.

Single imprinting is cueing fish to a single unique water supply prior to

transport and release. Various mechanical stimuli were used in combination

with the unique water source to attempt to achieve the single imprint. Most

fish subjected to a single imprint were transported by truck and released

below Bonneville Dam.

Sequential imprinting is cueing fish to two or more water sources in a

step-by-step process to establish a series of signposts for the route

"home". This was attempted in two ways. One method was to change the water

supply for the fish at the hatchery. The second method was to use a barge

with a regulated flow-through water system. Various test lots of fish were

then transported by truck, truck and barge, or barge only to a lower river

release site.

With variations, these three techniques were used for all species tested:

coho salmon, spring and fall chinook salmon, and steelhead. because of the

limited availability of hatchery fish, a number of different racial stocks

within the various species were used in the experiments.

11

The degree of success (ability to home and survival enhancement) for the

various treatments of experimental fish is based on the returns of adults

previously marked as juveniles with a CWT. Homing is determined by the rate

of return of marked adults to the homing sites, and relative survival is

measured by the combined total recoveries of CWT at the homing site, from

in-river sampling sites, from ocean and river commercial and sport fisheries,

and from hatcheries and spawning grounds. Discrete multivariate analysis was

used to statistically compare test and control treatments of completed

experiments (Bishop et al. 1975). In this procedure, the treatments were

structured in contingency tables. The C-statistic was used to test for

significance which was established at P < 0.05, df = 1 (Sokal and Kohlf 1981).

The study began in 1978,, and the juvenile marking phase was completed in

1 9 8 0 . During the J-year marking phase, over 4 million juvenile salmon and

steelhead were marked and released in 2 3 experiments. The NMFS conducted 19

of the experiments and the Idaho Cooperative Fishery Research Unit (ICFRU)

conducted four under contract to NMFS (Table 1). The 6 years of activities

and initial results from the 23 experiments were previously reported by

Slatick et al. (1979, 1980, 1981a, 1981b, 1982, 1983, 1984) and in Volume III,

Novotny and Zaugg (1979, 1981, 1984). Keturns of adult coho salmon, spring

and fall chinook salmon, and steelhead are now complete. The final analysis

of results are presented by species in this report.

COHO SALMON EXPERIMENTS

Juvenile coho salmon used in all homing experiments were progeny of

adults which return to the Little White Salmon Kiver. This stock of coho

salmon originated in the Toutle Kiver. Little White Salmon and Willard

12

Table l.--Homing imprint experiments l978-80--species, location, numbers of fishmarked and released.

---------- ___------

Species and hatchery ofYear fish marked and released1978 197Y 1980

origin-homing site (no. ) (no. ) (no. )

SteelheadDworsnakTucannonTucannon-L. Goose Dam

Spring chinook salmonKooskiaRapid River

Fall chinook salmonHagerman-Lower Granite Dam

SteelheadChelan-LeavenworthWells-Winthrop

Spring chinook salmonCarson-PascoCarsonLeavenworth

Coho salmonCarson-PascoWillard-Stavebolt Cr.Willard

Fall chinook salmonWhite Salmon-Stavebolt Cr.Big Creek-Stavebolt CreekSpring Creek

Snake RIver System

74,741 99,12536,686 67,573

78,091

186,597 123,600121,566

114,000

Columbia giver System

137,949 137,81796,978 62,243

113,681159,682 159,327

491,768

102,594414,907

436,118

473,027143,605259, 786

Subtotals by species

Spring chinook salmonFall chinook salmonCoho salmonSteelhead

Grand Total

186,597 273,363 896,261 1,356,221473,027 517,591 990,618

517,501 436,118 953,619346,354 270,633 183,227 794,213

1,050,452 1,017,023 2,027,196 4,094,671

13

National Fish Hatcheries (NFH) are located on the Little White Salmon River,

which provides the water supply for both hatcheries. All returning adult coho

salmon are collected at the Little White Salmon NFH which is located near the

mouth of the river. Juveniles were reared at the Willard NFH, 3.5 miles

upstream.

Background

Carson NFH - Pasco Homing Site, 1978

This experiment was concerned with imprinting juvenile coho salmon reared

at a mid-river hatchery to home as adults to an upriver homing site. Juveni le

coho salmon reared at Carson NFH (Little White Salmon brood) were subjected to

single, sequential, and natural imprinting at the Pasco homing facility. The

objectives of the experiment were to:

1. Determine the relative effectiveness of the three imprinting

techniques for returning adults to the Pasco facility.

2. Define the effect of the three imprinting techniques on overall

survival.

Experimental Design

Juvenile coho salmon were reared in raceways and pre-marked at Carson

NFH. Three groups of fish were imprinted for MINIMUN of 48 h at the Pasco

homing site. The naturally imprinted group (43,961) was released at the

homing site. The single imprint group (28,927) was trucked to below

Bonnevi L LE Dam. The sequentially imprinted group (29,706) was trucked to a

barge at Richland, Washington, and barged below Bonneville Dam (see Volume II,

Table Al.0 for additional details). Adult returns to the Pasco facility,

inriver sampling sites, and the various fisheries were used to evaluate the

effectiveness of the three imprinting techniques.

14

Kesults and Discussion

A total of 6 jack and 10 adult coho salmon from the naturally imprinted

release returned to the Pasco homing site; no fish from the other groups

returned to the homing site. Nine coho salmon were recovered at river sampling

sites: five from the naturally imprinted group, one each from the barged and

the trucked groups at Bonneville Dam, and two fron the barged group at McNary

Dam. Based on these returns, it is apparent that the coho salmon used in the

truck (single imprint) or barge (sequential imprint) experiments did not

receive an adequate homing imprint to the Pasco water supply.

A total of 116 tags were recovered from the ocean commercial and sport

fishery, and 21 tags were recovered from the Colmbia River gillnet fishery,

Zones l-6 (Table 2). Survival cf fish from the naturally imprinted group was

significantly greater than for fish from the other groups (P < 0.01

df = 1). The naturally imprinted group contributed over twice as many fish to

the fishery as either of the other groups.

The poor contribution of the single and sequentially imprinted fish to

the various fisheries probably resulted from poor survival of the juveniles

after their release below Bonneville Dam. Sampling of juveniles in the

estuary at Jones Beach (KM 47) also indicated that survival of the naturally

imprinted fish was twice that of the single and sequentially imprinted fish

(Dawley et al. 1 9 7 9 Cause of the apparent mortality is unknown; all fish

appeared in good condition at time of release. However, a factor which may

have influenced the lower survival of the fish released below Bonneville Dam

could have been the latent effects of pathogenic infections combined with

stress induced by the added crowding, handling, and transport during the

experimental releases. A few weeks before the fish were move to Pasco they

Table 2 .--A camparison between recoveries of natural, single, and sequentially imprinted groups of adult coho salmon from marked groups ofjuveniles reared at the Carson Hatchery and imprinted to the Pasco homing site. Recoveries were from September 1978 to February1980.

Juveniles released Adult returnsAdult traps STRAY Pasco Total

COMMERCIAL & sport fishery Bonneville 6 to homlng combinedOcean River combined McNary Dams hatcheries site return

location NO. Date (no.) (no.) No. % T/C ratio (no.) (no.) no. No. % T/C ratio

pasconatural 43,961 03 May 75 13 88lmprlnt

Bonneville

single 26,927 01 May 24 5 29Imprint

sequential 29,706 04 May 17 3 20- -Imprint

Total 102,594 116 21 137 9 5 16 167

0.200 5 1 16 110 0.250

0.100 0.50:1 1 3 0 33 0.114 0 . 4 6 : 1 15v,

0.067 0.34:1 3 1 2 4 0.081 0 . 3 2 : 1

l * = significant difference between single and sequentially imprinted groups and the naturally imprinted group (P < 0.01, df n 1).

16

experienced an outbreak of coldwater disease and incurred a mortality of 7.8%

Leek2/).

Conclusions

1. Natural imprinting was successful in returning adult coho salmon to

the Pasco homing site (164 miles upriver from the hatchery of origin).

2. The single and sequential imprint methods were unsuccessful in

providing a homing cue in coho salmon which would enable adults to return to

the homing site.

3. Survival of fish from the natural imprint group was over twice that

of the single and sequential imprint groups.

2/ Steve Leek, Little White Salmon NFH, Cook, Washingtoncommun. 1979.

98605, pers.

17

Willard NFH - Stavebolt Creek, 1978

Background

This experiment was concerned with imprinting juvenile coho salmon reared

at a mid-river site to home as adults to a lower river site. Juveniles reared

at Willard NFH were subjected to natural and single imprinting at the

Stavebolt Creek homing site (Fig. 1). Success was evaluated by comparing

return rates to Stavebolt Creek (tests) with return rates to Willard NFH

(controls). The objectives of the experiment were to:

1. Determine the effectiveness of single and natural imprinting of

juvenile coho salmon in returning adults to a lower river homing site

downstream from their hatchery of origin.

2. Determine the length of time required to imprint a specific homing

cue.

3. Define the effect of the level of smoltification on the ability of

juvenile coho salmon to accept a specific homing cue.

The Stavebolt Creek homing site, located in northwestern Oregon, was

selected for its natural features and made available through the generosity of

Mr. Carl Utzinger in allowing us to use the site over the period of our

experiments. Stavebolt Creek is a small tributary to the Lewis and Clark

River, one of three river systems flowing into Youngs Bay near Astoria, Oregon

(Fig. 4). The homing site consisted of: (1) a pond supplied solely by

Stavebolt Creek water where fish were held in floating net-pens for

imprinting; (2) 600 feet of creek, from the pond to its confluence with the

Lewis and Clark River, for natural migration imprinting; and (3) a fish trap

near the mouth of the creek for recovering returning adults.

18

YOUNGS BAY DRAINAGEClatsop County. Oregon

0 Creek tr8p site8 Stnycohoaptunsi9- Stream surwys

/

- Fish passage barriers;.$:.z fwdd fishing l m

* Hammcmd relmsa sitet

Miles, * 1012345

Figure 4.--Location map of release sites and recovery areas pertainingto imprinting coho salmon in the lower Columbia River area.

19

lE.

1ck>=0t

5+Ytz

5

0

TEST SERIES1St 2nd 3rd

I I I I I I I I

6 13 20 27 3 10 17 24 1 8 15 22

\

\88888

29 5

\

I I

12 19 26

0.15

0.10 H8kSa.iiB

0.05 f

0

Figure 5.-Na+-K+ ATPase activity profile for coho salmon reared at WillardNFH indicating time frame for releasing imprinted fish for lst,2nd, and 3rd replicates in 1978. This figure also illustratesthe correlation between recoveries of adult coho salmon at theLittle White Salmon NFH and in THEOCEAN and Columbia Riverfisheries in relation to the Na -K ATPase activity profile ofthe jwenile fish released as controls from the Willard NFH.Recoveries are from September 1978 to March 1980.

20

Since our homing strategy is based on olfactory perception of unique

water sources, a desirable feature of the Stavebolt Creek homing site was that

the migratory route (north to south) for adults was directly opposite the

route from Willard NFH (south to north). This directional difference of the

migration route to Stavebolt Creek should not influence returning adults if

the homing imprint used by salmon was olfactory perception of a scent route to

the homing imprint site. To examine if a single exposure to a unique water

source would implant a homing imprint in juvenile coho salmon, the fish were

hauled out of the Youngs Bay watershed and released into the Columbia River at

Hammond, Oregon (Fig. 5).

Experimental Design

Juvenile coho salmon were reared and pre-marked at the Willard NFH.

Na+-K+ ATPase enzyme activity was monitored at the hatchery, and experimental

and control releases of marked fish were designed to coincide with rising,

peak, and falling N A + K + ATPase enzyme activity (Fig. 5). Each release series

consisted of six test groups and one control group of approximately 20,000

fish each. All control groups were released at Willard NFH (natural

imprint). During the rising and the beginning of the decline of the Na+-K+

ATPase activity, groups of test fish were trucked to Stavebolt Creek (Fig. 4),

held in live pens for various periods of time, and released either into

Stavebolt Creek directly (natural imprint) or trucked to Hammond, Oregon

(RM 8), and released into the Columbia River (single imprint).

During the third release series (falling Na+-K+ ATPase activity), the

experimental design was aborted due to high water temperatures in the

Stavebolt Creek holding pond. An alternate experiment was substituted for the

four marked groups remaining at the Willard NFH. The control release was made

at Willard NFH as before. The three test groups already at the Stavebolt

21

facility were released into Stavebolt Creek (natural imprint). TWO of the

three test groups remaining at Willard NFH were placed in tank trucks for 2 h,

then returned to the raceway for 4 h; one group was then trucked to below

Bonneville Dam and released, and the second group was trucked to Hammond,

Oregon, and released (single imprint). The third test group was trucked

directly from the hatchery to the release site below Bonneville Dam (no

imprint). The objective of this alternative experiment was to determine if a

stimulus like simulated trucking could be used to imprint fish transported and

released at other sites to home to their hatchery of origin (see Volume II,

Table A2.0 for additional details on experimental design).

Results and Discussion

Adult returns from single imprint tests indicate the methods used in 1978

were unsuccessful in returning coho salmon to either the Stavebolt Creek

homing site or Willard NFH. Four adults returned to Stavevolt Creek out of

80,000 juveniles imprinted for 24 to 48 h in Stavebolt Creek and then

transported and released at Hammond, Oregon. No fish returned from the 4- to

12-h imprint groups. By contrast, 26 fish returned to Willard NFH out of

40,000 juveniles released there as controls (Table 3).

Results from sampling the Columbia River and Youngs Bay fisheries

indicated that although the single imprint method used in these experiments

was not successful in returning fish to the homing site, it did implant a

limited homing cue which caused coho salmon to return to the geographic area

adjacent to the homing site, i.e., Youngs Bay (Table 4). This homing cue to

Youngs Bay may have been reinforced by the location of the Hammond release

site. The flow of the Columbia River would have carried water exiting Youngs

Bay along the south shore past the Hammond release site (Fig. 4).

22

Table 3 .-A ccznparison bet;reen control and test groups of adult coho salmon returning to Little White

Salmon NFH and Stavebolt Creek homing sites and recovered in the ocean and river fisheries. Recoveries are from September 1978 to December 1979.

HS31!?(! imprint

No. of grows

Juveniles released ---. .___ ._. ._- -..---. -...-

Adults recovered Hominq sites Ocean and river fisheries

SO. Location - Date .-- No. % No. x TIC ratio

c3!Ttr3!.

Single 4 6 12 h

24 S 48 h

Natural 4h

Control

Sinqle 4 b 12 h

24 6 48 h

Natural 4h

Contr 01

Natural !?T-iT-

tie inprint

Single 4 h

4 h

Na+-K+ ATPase, 1st Release

1 19,033 Hatchery

2 39,364 Hammond 2 40,280 Hammond

2 40,245 Stavebolt

Na+-K+ ATPase, 2nd Release

1 19,943 Hatchery

2 39,354 Hammond 2 39,832 Eaznond

2 41,555 Stavebolt

Na+-K+ ATPase 3rd Release

1 19,781 Hatchery

2 37,857 Stavebolt

1 19,771 Below Bonneville

1 19,730 Below Bonneville

09 Yay 8

12 b

13 Jun 0 10 6 11 Jun 3

11 6 12 Hay 15

0.040 17

0.000 -- 51

0.007 l 37

0.037 NS 54

0.085

0.130

0.092

1.53:1 NS

1.08:1 ss

0.134 1.58:1 9s

24 Hay 18

24 i 25 May 0

25 6 26 Hap 1

25 Hay 15

0.090 20

0.000 -- 42

0.003 - 44

0.036 * 52

0.100

0.105

0.110

1.95:l US

l.lO:l ss

0.125 1.25:1 NS

08 Jun 9 0.045 16

07 Jun 16 0.042 NS 41

08 Jun 0 0.000 - 29

0.081

0.108 1.33:l ss

0.147 1.81~1 NS

08 Jun 1 0.005 -- 27 0.137 1.69:1 NS

1 19,622 Hammond 08 Jun 0 0.000 -- 19 ..- _. _ 0.097 1.2O:l ss

t = Significant difference between test and control group (P < 0.05, df - 1). NS = Nonsignificant. -- s Insufficient recoveries for statistical analysis.

Table 40-A comparison between control and the groups of coho salmon recovered at Stavebolt Creek,Little White Salmon Hatchery, commercial and sport fisheries, and as strays.

Adults recovered from group Imprinted as IndicatedStavebolt Creak L.W. Salmon Hatchery

Natural Single Slngle Slngle No

Recovery sitesControl

o/IMPRINT IMPRINT IMPRINT IMPRINT

L.W. Salmon (Stavabolt)w (HammondIw (Hanunond# (B'vlle.@

Homing SITES

Stavebolt CREEK 0L.W. SALMON Hatchery 33

Fisheries

46 4 0 0 0 500 0 0 1 0 36

Youngr B a y 2 15 30 1 0 0 48Zone 1-2 0 12 12 4 1 1 SOZone 6 4 0 0 0 1 2 7Ocean 47 120 132 14 25 26 #4

Total 88 197 183 19 29 31 547

%! 59,632 smolts released.ldCl

119,657 smolts released.` 159,330 smolts released.2 19,622 smolts released.

7..19,730 smolts released.

- 19,771 smolts released.a/ 397,742 smolts released.

24

Adults returning from juveniles released at Stavebolt Creek (natural

imprint) showed a positive homing response to Stavebolt Creek. Adults from

the 1st and 3rd Na+-K+ ATPase release returned to the Stavebolt Creek homing

site at about the same rate as controls returned to the hatchery (Table 3).

Straying of adult coho salmon from the Stavebolt Creek (natural imprint)

and Hammond (single imprint) releases to other streams in the Youngs Bay

system was negligible. One fish was recovered in Hartell Creek, a tributary

to the Lewis and Clark River, located 1 mile upstream from the Stavebolt Creek

homing site. N o marked fish were recovered in the other four stream systems

draining into Youngs Bay and containing spawning coho salmon (Fig. 4).

Survival [baaed on test/control (T/C) ratios] of fish recovered in the

ocean and river fisheries that were given a single imprint was slightly better

than survival of those released at the 'hatchery, but the difference in the T/C

ratios was not significant (Table 3). There appeared to be little difference

between the single and natural imprint methods in relation to relative

survival (Table A), but more fish from the Hammond release were captured in

the Youngs gay fishery (Table 4).

As suggested by Vreeland et al. 1 9 7 5 one obvious application of using

a technique that gives a homing cue is to provide a site specific fishery. As

an example, fish surplus to an up-river hatchery's need could be transported

to Youngs Day, given a natural homing imprint, and released. This could

provide more adults to the riverine fishery and provide an area where fish

could be efficiently harvested without impacting other runs of fish.

An attempt was made to make releases of coho salmon coinciding with

rising, high, and declining NA+ k + ATPase enzyme activity to determine its

potential as an indicator of the best time period to imprint a homing cue.

Data from adult returns of those fish released as controls at the hatchery

25

show that the 2nd release had the best recovery rate both back at the hatchery

and in the fishery (Fig. 5). Returns to the hatchery of fish from the 2nd

release were statistically significantly better than returns from the 1st and

3rd releases (P < 0.05, df = 1). Whether this was a function of time of

release (9 May, 24 May, or 8 June) or enzyme activity is not known. The rates

of return of test fish to homing sites and in the fishery for all three

release periods, however, were about the same, indicating time of release or

enzyme activitiy did not have any effect on homing of the various test

treatment groups released in 1978 (Table 3).

Based on recoveries from the alternative test durfng the 3rd release

series, releasing coho salmon below Bonneville Dam appeared to increase their

contribution to the various fisheries by 69 to 81%,, but these fish failed to

home back to their hatchery of origin (Table 3). Additional detail on returns

from specific releases may be found in Volume II, Tables A2.1 to A2.21.

Conclusions

1. The single imprint method used in 1978 was unsuccessful in enabling

adult coho salmon to home back to either Stavebolt Creek or to the Little

White Salmon homing sites.

2. Adults which had received a single imprint to Stavebolt Creek

(Hammond release) generally homed to Youngs Bay, exhibittng a limited or

partial homing response.

3. Coho salmon imprinted and released in Stavebolt Creek (natural

imprint) in the 1st and 3rd release series returned to that homing site at

about the same rate as control releases back to the hatchery.

26

4. Survival of both single and natural imprint fish recovered in the

ocean and river fisheries was higher but not significantly higher than those

released at the hatchery.

5. Imprint techniques used in this experiment can provide fishery

managers with a site specific fishery (Youngs Bay) and lower Columbia River

while still providing sufficient fish for spawning.

6. Of the three control releases, the 2nd release (near the beginning of

the decline of the Na+-K+ ATPase curve) had the highest catch rate in the

fishery and the highest return rate to the hatchery.

7. No optimum homing imprint period was noted between the three series

of test releases of coho salmon which had been imprinted to Stavebolt Creek.

8. Straying of fish from the test groups imprinted to Stavebolt Creek

within the Youngs Bay system was negligible.

27

Background

Willard NFH, 1980

The goal of this experiment was to imprint juvenile coho salmon which

were to be transported by truck and barge from their hatchery of origin and

released at various locations below Bonneville Dam to return as adults to the

Little White Salmon giver (hatchery of origin). Juveniles reared at Willard

NFH were subjected to single and sequential imprinting methods before being

transported. The adult returns from these test groups were compared with

adult returns from fish which received a natural migration imprint during

their outmigration. The study was designed to determine:

1. Effectiveness of various methods used to activate a homing imprint in

coho salmon.

2. Effect of various release locations on the homing ability and

relative survival of coho salmon.

3. Effect on relative survival of fish marked in the fall as juveniles

vs fish marked as smolting fish in the spring.

Willard NFH is part of the Little White Salmon-Willard Hatchery complex

operated by the U.S. Fish and Wildlife Service (USFWS) and located on the

Little White Salmon giver in southwestern Washington (Fig. 6). Coho salmon

released at Willard NFH migrate through 3.5 miles of free-flowing river before

entering slack water at Drano Lake. Waters from the Little White Salmon River

remain distinct in Drano Lake before merging with the Columbia River at RM

162.0.

A barrier-dam and fish collection facility a few hundred feet above slack

water at the Little White Salmon N F H blocks access of returning adults to

Willard NFH All adult coho salmon returning from Willard N F H releases are

collected and held for brood stock at Little White Salmon NFH.

28

OREGON\AREA l o c a t i o n

0 8Reofan¶trolreio8sen Hetdwry1 Bonneville Dam Rm 1402 Beaver Terminal Rm 503 Hammond RM 8

Willard NationalFish Hatchery 4

Little White SalmonNational Fish Hatchery

’ Lake ) /

Figure 6.--Little White Salmon - Willard National Fish Hatchery complexand transport routes, 1980.

29

The site was conducive to technical requirements of the study,

particularly the recapture of juvenile migrants. The capability of mooring a

fish transport barge in the Little White Salmon River arm of Drano Lake within

200 yards of the Little White Salmon NFH was also an important

consideration. A timely evaluation of the study results would be possible,

since virtually all rack recoveries of adult coho salmon reared at Willard NFH

are completed within approximately 16 months after their release as smolts.

Experimental Design

Experimental groups were provided imprint cues to Little White Salmon

River water and then transported by barge or truck to release sites in the

Columbia River below Bonneville Dam (RM 140 and 142, respectively), or by

truck to Beaver Terminal, Oregon (RM 50), or by truck to Hammond, Oregon (RM

8). The barging portion (sequential imprint) consisted of three test groups

that received different treatments in an effort to imprint the smolts to water

from the Little White Salmon River. A prior event group (Test Group 1) was

passed through 175 feet of pipe into a lower raceway to subject the smolts to

an “event” which might possibly trigger their natural homing imprint

mechanism. To approximate the natural situation as closely as possible, fish

for the limited migration group (Test Group 2) were recaptured at the Little

White Salmon NFH following a natural migration of 3.5 miles from Willard

NFH. The barged only group (Test Group 3) received no artificial stimulation

other than being transported by truck from the raceway they were reared in at

Willard NFH to the barge moored in Drano Lake. Three test releases received a

single imprint simulated truck release) and were trucked from the hatchery to

three sites below Bonneville Dam (see Volume 11, Table A3.0 for additional

detail on experimental design).

30

Recapture of juvenile migrants from the Little White Salmon River was

attempted using a self-cleaning scoop trap (Raymond and Collins 1974), but

trap efficiency was too low to supply the 50,000-fish goal for Test Group 2

(limited nigration). Migrants which entered the water intake to the adult

holding ponds at Little White Salmon NFH were captured and used to supplement

the trap catch. Only 33,372 juveniles were released--a smaller than desired

group.

handling and especially marking smolted salmonids are generally

considered to cause decreased survival. The inclusion of Test Group 2

(recaptured natural migrants) in the study design made it necessary to mark

this group during the smolting period. To avoid bias, other groups were also

marked in the spring. Concern over the possible adverse effect of spring

marking led to the inclusion of Control Group 2 which was marked in November

1979, Their survival was compared with Control Grouip 3 marked in the spring

(both were released in the Little White Salmon River on 23 Hay).


Statistical analysis of hatchery and ocean recoveries determined there

was no significant difference (P < 0.05, df = 1) between Control Group 2 (fall

markedj and Control Group 3 (spring marked) recovered in the ocean or at the

hatchery (Comparison 1, Table 5). Since there was no significant difference

between Control Groups 2 and 3 , they were combined to strengthen the

statistical analysis.

Homing of the barged groups to the hatchery was quite effective as

indicated by only a 0.13 difference between the T/C ratios in the ocean and at

the homing site 0.69:1 -and 0.56:1, respectively) (Table 6). Most of this

0.13 differential in homing ability was accounted for in increased

contribution to the Indian fishery and strays into other hatcheries in the

31

Table 5 .--Statistical treatment of Willard NFH coho salmon homing experiment.

ComparisonRecovery area

Ocean Hatchery

1.2.3.4.5.

6.7.8.

9.

10.11.

Control 2 vss Control 3Control 1 V8s Control 2 & 3Barge Test 1 vs 2 vs 3Truck Test 5 vs 6Pooled barge (Tests 1, 2, 3) vs pooled truck

(Tests 5 & 6)Pooled truck (Test 5 & 6) vs Truck Test 4Pooled barge (Tests 1, 2, 3) vss Truck Test 4Pooled barge (Tests 1, 2, 3) vs pooled control

(Groups 2 i 3)Pooled truck (Tests 5 & 6) vs pooled control

(Tests 2 & 3)Pooled truck (Tests 5 & 6) vs Control 1Pooled barge (Tests 1, 2, 3) vss Control 1

NS NS* *NS NSNS --

NS --* --* --

*

**l

*

----*

* = Significant difference between test and control releases (P < 0.05, df = 1).NS = Nonsignificant.-- = No test.

32

Tabk 6.--Coqrlsm bmtmmm adrot omItat-sof drli- ,.lra rrarrd ut tbu Llttl. unlh 9.14 m an4

In moocrn Oshul.¶ fCQ r.I.m¶osQf t91ts f~*rllldwtlwhleh - Ilylntad fu thml.IttI~Ilhlh

Sal=lm Rlrrr ad c.l..s.4 at 618 6lfforoat ,I* In 1990. ~lr~ttraJI26F~l9g.

krnnlles rel.a.4 M9ltrmrlr -

Qm

lrutmnt Slh 09t. Id wa "7 l/c rutlo IIBO 1 TIC rat10

cxmfrol -s fn6tural Isrlntl

eYtrol I Fall rr* L.Y. Smlmn Kw

bdrol2 Fdlmuk WIllad m

GJmtrol Jblnp Ir* Ulllwd )(Fw

bntrolZb3

fpaolo6l Wlllw4 m

Ru9md 9mws inwant Id ImInt)

Tat1 hla - -Ill9 (RI 140)

Tat 2 Llmlt.6

mlgetlon Ibnnsrlll. w4 1401

rut 1 Rrgd only 9mmvlll. ma 1401

lutr I, 2. L J

Cpoolo6d) 9mlmllk (RI 140)

~mb3d q-am6 CSlngl~ Iqrlni)

Tut 4 SId&yl,

14 lfuy 23 -v 23 wrr

25 f-v

25 w =w

25 -v

r.l.dY OmltoaPolmt (RI I42l 2IW Ted 5 Slmulatd

rdeesa Y (rn 0) 23 law Tut 6 slmolstd

+dMS. ewwr Temlnd

(RI1 SOB1 22-v

7uts’ILd

wooldl l3cmy ar.a

4S,O45

42.nl

¶1,523

Yl.417

33.m

47,293

lS3.022

50,766

50.619

51,693

102.sO2

40

109

l4Y

25s

n

47

19

201

7

0

0

0

0.09s

0.2%

0.291

0.269

0.146

0.139

0.169

O.lYl

0.014

0.000

0.000

o.ooo

9.59:1

0.92: I

0.62: I

0.56: I

0.06: I

0.69~1

0.64: 1

0.7s: I

O&9:1

0.45: I

017l:l

0.66: I

0.m I

g MJntd far t69 loss.

t? l.oaded In hick far 2 II ttan r.l.msd Into ru0w.y umtmlmlmg L.U. s.lmom Rlrr uoiur w6tu fff 49 )I l lnlu nl.a tr6m666rtd by

track omtalrlnp L. w. salon wofwr.

33

Bonneville Dam area (Table 7). If the Indian fishery (Zone 6) and stray fish

recoveries are added to the numbers of fish which returned to the homing site,

the T/C ratio of adults which returned to the Bonneville area from the barged

groups was approximately the same as in the ocean 0.68:1 and 0.69:1,

respectively). The data further indicated that when imprinting coho salmon

smolts to the Little White Salmon River, the direct truck to barging process

alone was reasonably effective, and additional stimulation or a short natural

migration was not necessary (Comparison 3, Table 5--no significant differences

between recoveries of the three barge treatments in either the ocean or back

to the hatchery).

The positive homing by the barged groups must be qualified by the

following: (1) the return of adult coho salmon to the hatchery from the

natural migration (control) releases (Coaparison 8, Table 5) was significantly

(p < 0.01, df = 1) greater than the return of fish from the barged groups and

(2) the increased rate of recovery in the Indian fishery and higher straying

rate of fish from the barged groups than from the control groups (Table 7)

indicate that some barged fish exhibited a honing impairment to the Little

White Salmon giver. Howeve r, the majority of returning adults from both the

barged and control groups honed back to the Bonneville area.

By contrast, the single imprint method (direct trucking from the

hatchery) used in this experiment did not adequately imprint the juvenile coho

salmon to home successfully as adults to the hatchery (homing site). None of

the fish trucked to and released at Beaver Terminal and Hammond, Oregon,

returned to the hatchery (Table 6). Instead, these fish returned to the

release site area as indicated by 56 recoveries in the lower river fishery

(Zone 1-5 and Youngs Bay) compared to no recoveries above Bonneville D a m

either in the Indian (Zone 6) fishery or the hatcheries (Table 7). Fish

34

Table 7. -A comparison between recoveries of control and test groups of adultcoho salmon from the 1980 Willard NFH experiment in variousfisheries and as strays to hatcheries in the Columbia River system.

Pooled Pooled Pooled Dalton Point

Site -_ - ____ _ _ c o n t r o l s barged trucked trucked---- -

Fisheries- -

Zone l-5Youngs BayZone 6Washington terminalSport fishery

11

1740

3422000

Hatcheries

Bonneville

Cascade

130-

15110

2 60 101 1- -

Total all areas 9 49 58 21

alb /

Skamakawa Creek, Cowlitz River, and Grays River.Grays River and Washougal SFH.

35

released at Dalton Point, Oregon, returned as adults to the Bonneville Dam

area; however, from this location they strayed into various hatcheries in the

area, including the Little White Salmon NFH (Tables 6 and 7).

Relative survival as measured by recovery of adults in ocean fisheries

indicates no significant differences between the barged groups, the trucked

groups released at Beaver Terminal and Hammond or these groups combined

(Comparisons 3, 4, and 5, Table 5). However , both the barged and lower river

truck release groups contributed significantly (P < 0.05, df = 1) more fish

(51%) to the ocean fisheries than did the Dalton Point truck release group

(Table 6). The increased contribution to the ocean fisheries of the barged

releases (RM 140) over the Dalton Point shore release (RM 142) in basically

the same area suggests that a mid-river release in the main channel may be

more productive than the shore release site. However, the eruption of Mount

St. Helens could also have been a factor inn the lower survival of the Dalton

Point release (discussed later). Additional details on recoveries from

specific releases may be found in Volume II, Tables A3.1 to A3.9).

Survival of the pooled controls (Groups 2 and 3) was significantly

greater than survival of either the barged or trucked test groups (Comparisons

8 and 9, Table 5) (P < 0.05, df = 1). The overall rate of return for the

pooled controls was 0.3% vs about 0.2% for the transported groups (Table 6).

This was unexpected, since the Stavebolt Greek-1978 study (discussed

previously) and other studies (Ebel 1970; Slatick et al. 1980; McCabe et al.

1983) demonstrated equal or better survival for fish transported and released

below Bonneville Dam compared to fish released at the hatchery. Data on

returning adult fish from fall chinook salmon released below Bonneville Dam in

1979 and 1980 (discussed later in this report) also indicate better survival

of transported fish.

36

The return rate for the first control release (0.1%) (Table 6) was

significantly lower than either of the other two control groups (Comparison 2,

Table 5) or the transported groups (Conparison 10 and 11, Table 5) (P < 0.05,

df = 1). Possible reasons for the poor returns of the transport groups and

the first control group include: (i) stress placed on fish during handling,

marking, loading, and transporting; (2) bias from different quality fish

between raceways (fish were not randomized prior to marking); and/or (3) the

eruption of Mount St. Helens.

With respect to potential stress, there are several factors to

consider. Control Group 1 was transported in a 1,500-gallon hatchery truck

from Willard NFH and released below Little White Salmon N F H whereas the other

two controls (fall vs spring marking comparisons) were released at Willard N F H

without added handling or transportation. T w o of the barged groups were

handled and marked within 5 days of release. The third barge group, as well

as the truck releases, were marked approximately 1 month prior to release. It

is possible that the added stress of crowding, loading, and transporting

shortly after marking could impact survival. Seawater challenge tests for

measuring stress indicated that stress levels of handled and marked fish are

significantly higher than the stress levels of unmarked fish (Park et al.

1982.).

Differences in rate of return between transported and control fish could

have resulted if the quality of fish varied significantly between raceways.

The experimental design made it nearly impossible to randomize fish prior to

marking. The NMFS did request though, that the fish be comparable in size and

weight and be representative of the production release.

Mount St. Helens erupted on 18 Ma y 1980, and the subsequent peak runoff

of suspended solids affecting the Columbia Kiver was in evidence by 19 May.

37

Control Croup 1 inadvertently released on 14 May may have been extremely

impacted by the relatively warm, turbid flows in the vicinity of the

confluence of the Cowlitz giver ( R M 6 8 ) Data from NMPS sampling programs

indicate that juveniles from the first control release reached Jones Beach

(RM 47) on 19 May, coincident with the peak runoff from the eruption (Dawley

et al. 1981). In contrast, the pooled control group (Groups 2 and 3) arrived

at Jones Beach around 1 June, after the river conditions had significantly

improved.

Nount St. Helens may also have impacted the test groups. The barged fish

(Test Groups 1, 2, and 3) were released below Bonneville Dam on 25 May. Test

Group 4 (Dalton Point), with the lower survival, was released on 21 May, 4

days earlier. Tests Groups 5 and d (Hammond and Beaver Terminal) were trucked

downstream and released on 22 and 23 I-lay, respectively, directly into Columbia

Kiver water impacted by Mount St. Helens effluent. Timing and location of the

release appears critical. The high water temperatures and turbidity from the

eruption lasted only a few days. T h e ‘Beaver Terminal (RM 50) and Hammond

(RM 8) releases were on the south shore where the main flow of the Columbia

River would have diluted the deleterious effects of the volcanic plume to a

degree. However fish from the Dalton Point (RM 14.L) release may have been

more randomly dispersed across the Columbia giver when they came into contact

with the more concentrated volcanic plume from the Cowlitz River (RM 68) on

the north shore. Therefore,, a greater number of fish from this release may

have been affected by the volcanic plume. There is evidence from Dawley et

al. 1 9 8 1 that juvenile salmonids migrating through the estuary shortly after

the eruption were adversely impacted by the poor environmental conditions

encountered. RWRE suspect the eruption of Mount St. Helens may have been the

major problem in the reduced survival of the transported fish in this

38

experiment. Similar stresses and non randomizing of fish in raceways have

occurred in previous experiments in which transported fish returned at a

higher rate than control fish.

Conclusions

1. There was no significant difference in adult survival between paired

releases of control groups of juveniles either marked as pre-smolts in the

fall or marked during their smolting period in the spring.

2. barged fish homed successfully to the Little White Salmon NFH. The

direct truck to barge process is adequate; no additional stimulation or short

natural migration appears necessary.

3. Fish trucked to and released at Beaver Terminal and Hammond, Oregon,

were imprinted to home to the lower river; none were recovered at the

hatchery.

4. Fish trucked to and released at Dalton Point, Oregon, homed back as

adults to the general vicinity of their release site (Bonneville area) and

then strayed to various hatcheries in the area.

5. Recoveries of adults from transported fish were significantly less

than recoveries of adults frou the control releases. We suspect the eruption

of Mount St. Helens may have been primarily responsible.

39

Overview of Imprint Treatments on Coho Salmon

Recoveries of adult coho salmon returnfng to the Columbia River system

indicated that a combination of imprint method, mode of transportation,

release site, timing, and physiological condition of the juveniles influenced

their future adult migratory behavior.

Natural Migration Imprint

As expected, the majority of the fish demonstrated a positive homing

response to their point of release--be it a hatchery or an upper or lower

river homing site. The rate of return to Stavebolt Creek was nearly as high

as the rate of return of the control releases to the hatchery.

Fish imprinted to areas above McNary Dam contributed more fish to the

upper river whereas those released in young Bay contributed substantially

more fish to the lower river fishery.

Single Exposure Imprint

The single imprint methods used in these experiments were unsuccessful in

returning adult coho salmon to the hatchery or to upper and lower river homing

sites. Those trucked from the Pasco homing site and released below Bonneville

Dam were generally recovered as adults in the Bonneville area, and those

imprinted to Stavebolt Creek and released at Hammond were recovered in the

Youngs Bay commercial fishery.

Sequential Exposure Imprint

The sequential imprint method employed with coho salmon in the 1980

Willard HFH experiment provided a positive homing imprint. Test to control

ratios of adult coho salmon from all three barged groups that returned to the

Little White Salmon River and vicinity were about the same as the T/C ratios

measured in the ocean. These returns suggest that the barging process from

40

the Little White Salmon River arm of Drano Lake provided an effective homing

cue. Prior stimulation or a short distance (limited) natural migration had no

effect on the degree of homing achieved by the barged test groups.

In contrast, the sequential imprint method used in 1978 did not supply

the homing cues necessary to return adult coho salmon to the Pasco homing

site. The return data did suggest that some fish acquired a homing imprint to

the Columbia river above McNary Dam when they were loaded and held in the

barge above Richland, Washington.

Application of Findings

1. Juvenile fish surplus to an upriver hatchery's need could be

transported to Youngs Bay, given a single exposure imprint, and released.

This should provide more adults overall by reducing dam-related mortalities to

smolts and providing an area where fish could be efficiently harvested without

impacting other runs of fish.

2. A similar fishery and enhanced survival could be realized along with

the means of providing sufficient fish for spawning by imprinting fish to

Stavebolt Creek (natural migration).

3. Surplus juvenile coho salmon could also be used to enhance the river

fisheries above Bonneville Dam. Smotls could be transported to a selected

upriver location (such as Richland, Washington), loaded into a barge for

imprinting, and barged down river to below Bonneville Dam to reduce d a m

related mortalities to smolts thereby enabling greater numbers of adults to be

available to the fisheries above Bonneville Dam.

41

SPRING CHINOOK SALMON EXPERIMENTS

Spring chinook salmon were used in five homing imprint studies by NMFS

and two by ICFRU. These chinook salmon represented two principal stocks which.

had been introduced to various hatcheries in the Columbia River Basin. The

majority of fish were of Carson NFH stock. This brood stock was initially

introduced to Carson NFH in 1955 and came from the heterogeneous population

passing Bonneville Dam enroute to various upriver tributary spawning areas

(Wahle and Cheney 1981; Zimmer et al. 1963). The second stock of chinook

salmon was from the Rapid River Hatchery in Idaho. This brood stock was

introduced to the hatchery in 1964 from the heterogenous population arriving

at Hells Canyon Dam on the Snake River (Levendofskel/). Details on the

treatments for each of the five experiments conducted by NMFS are contained in

Volume II, Table B1.O

Rackgroun d

Kooskia NFH, 1978

This experiment was concerned with imprinting juvenile spring chinook

salmon to be transported from an upriver hatchery and released at a downriver

site to return as adults to their hatchery of origin. Juveniles reared at

Kooskia NFH that were to be transported were subjected to single and

sequential imprinting techniques before being released into the Columbia River

below Bonneville Dam. Results from the test groups were compared to those of

fish released from the hatchery to migrate downstream naturally. The

objectives of the experiment were to:

3/ Tom Levendofske, Idaho Department of Fish and Came, Rapid River Hatchery,Riggins, Idaho, pers, comm. 1986.

42

1. Determine the effectiveness of single and sequential imprinting

techniques in returning adults to the hatchery in comparison to the normal

hatchery release method (natural migration imprinting).

2. Determine the effect of the natural, single, and sequentia 1

imprinting techniques on overall survival.

Experimental Design

Kooskia NFH water was the imprint media used to implant the primary

homing cue in the test fish. During this experiment, the hatchery water

supply was obtained from Clear Creek, a tributary of the Middle Fork of the

Clearwater River. The experimental design included a control group released

with the hatchery production fish into Clear Creek and four test groups that

received a single or sequential homing imprint. Juveniles in the single

imprint group were trucked in hatchery water to a release site below

Bonneville Dam; sequentially imprinted fish were transported by truck or truck

and barge to release sites below Bonneville Dam.

Results

Adult recoveries from the experiment were insufficient for analysis. A

summary of the adult recoveries by area is presented in Volume II, Tables B1.1

to B1.5.

Carson NFH, 1979 and 1980

Background


salmon to be transported from a mid-river area hatchery and released at

downriver sites to return- as adults to their hatchery of origin. Juveniles

reared at Carson NFH were subjected to single and sequential imprinting

techniques before being transported by truck and released at two sites in the

43

Columbia River below Bonneville Dam. Results from these test groups were

compared to those from fish released from the hatchery into the Wind River to

migrate downstream naturally. The study was &signed to determine:

1. The effectiveness of single and sequential imprinting techniques used

in returning adults to the hatchery in comparison to the normal hatchery

release method (natural migration imprinting).

2. The effect of various release locations on the relative overall

survival of spring chinook salmon.

Experimental Design

The experimental design consisted of a control group released from Carson

NFH and three test groups given a variatioa of the simulated release imprint

techniques. Following a simlated release, test groups were transported by

truck and released at Dalton Point (RM 142) or Hammond, Oregon (RM 8). All

fish were premarked several months prior to release.

Results

Adult recoveries from the experiments were insufficient for analysis. A

summary of adult recoveries by area is presented in Volume II, Tables B2.1 to

B2.8.

Background

Carson NFH - Pasco, 1979


salmon reared at a mid-river area hatchery to home as adults to an upriver

homing site. Juveniles reared at Carson NFH were subjected to single,

sequential, and natural imprinting techniques at the Pasco homing facility ( R M

326) before being released. The objectives of the experiment were to:

44

1. Determine the relative effectiveness of three imprinting techniques

in returning adults to the Pasco facility.

2. Define the effect of three imprinting techniques on overall survival.

exerimenta 1 Design

Spring chinook salmon for this experiment were reared in raceways and

pre-marked at the Carson NFH located in the Wind River drainage. The eggs for

these spring chinook salmon were obtained from adults returning naturally to

the Carson NFH.

Our experimental design consisted of a control group released at Pasco

and two test groups transported to below Bonneville Dam. The single imprint

group was transported by truck, and the sequentially imprinted group was

transported by a combination of truck and barge. The fish were released 21-28

April 1979. A similar experiment using coho salmon was conducted in 1978.

Results

Adult recoveries from the experiment were insufficient for analysis. A

summary of adult recoveries by area are presented in Volume II, Tables B3.1 to

B3.3.

Background

Leavenworth NFH, 1980


salmon reared at an upper mid-Columbia River hatchery and then transported and

released at various downriver sites to return as adults to their hatchery of

origin. Juveniles reared at Leavenworth NFH were imprinted to Icicle River

water by allowing them to migrate for 1 mile in the Icicle River bypass

channel or holding them in a pen before being transported by truck and

45

released into the Columbia giver at two downstream sites. The results of

these tests were compared with groups of fish released from the hatchery to

migrate downstream naturally. This study was designed to determine:

1. Effectiveness of the imprint techniques in returning adults to the

hatchery in comparison to the normal hatchery release method (natural

migration imprinting).

2. Effect of various release locations on relative overall survival.

3. The effect on relative survival of fish marked in the fall as

juveniles vs marking ssolting fish in the spring.

Experimental Design

Five marked groups of approximately 100,000 fish per group were used in

the study. With the exception of a control group marked in November 1979

experimental handling and marking took place during the spring of 1980

coincident with timing of the natural outmigration--a time we believed the

fish were most likely to accept imprinting and exhibit true volitional

migration. Handling of most marked groups was extensive. Experimental groups

requiring volitional migration were released at the head of the Icicle River

bypass channel, recaptured in an inclined screen trap, and then returned to

hatchery raceways for marking and subsequent transport.

Groups released at White Bluffs and Dalton Point were transported in

5,000-gallon tank trucks containing hatchery water. releases were made on

24 and 27 April and 1 May. For groups other than the fall-marked control,

fish released on different dates had unique cold brands and wire tag codes.

This procedure allowed evaluation of returns in the event of significant

mortality in an individual transport load.

46


Statistical analysis of homing objectives was not possible due to low

returns for the SPRIG-MARKED control and truck transport GROUPS. Although

returns were low, it is noteworthy that fish transported to White Bluffs (RM

362) returned to Leavenworth NFH about as well as fish from the spring-marked

control group (Table 8). There were no returns to the hatchery from fish

transported to Dalton Point (RM 142) either from the volitional-migrant or

pen-held group, and of the five observed returns, three were indicative of

strayfng (recoveries in the Drano Lake and Sherears Falls sport fisheries and

at Klickitat SFH). A summary of adult recoveries by area is presented in

Volume II, Tables B4.1 to B4.5.

Homing behavior shown by fish from the White Bluffs release May have

resulted from imprinting acquired during migration down the Icicle River

bypass channel. Lack of homing for the corresponding Dalton Point releases

indicates that regardless of source, the imprint was insufficient to guide the

return of fish which had been transported as far downstream as Dalton Point.

Migrating juveniles were sampled at McNary and John Day dams and in the

lower Columbia River and at Jones Beach (Dawley et al. 1981; Sims et al.

1981). The data indicated higher in-river juvenile survival for fish

transported to White Bluffs and Dalton Point than for control releases from

Leavenworth NFH (Table 8). Survival of spring and fall marked control groups

was about equal.

To provide data regarding the effect of transport stress on survival,

NMFS personnel met each of the six Dalton Point transport loads, removed

samples of approximately 200 fish, and held the samples for observation of

47

fable R.--Secoverles of SprinE chinook salmon marked for the 1980 Leavenworth NFH homing experiment.

Experizental group and (number released) Control I Control 2 Test 1 Test 2 Test 3

marked fall marked spr “g volitional mfgr. volitional mfpr. pen hel+ 1070 1980 White Bluffs releaee Dalton Pt. release Dalton Pt. rn1p.t~~

Recoverv area (35,538) (98,7R9) (100,105~ (9FI.448) (96.663’)

Juvenile recoveries

a/ UC’:R rv na-- John bav n.3 ai

3 Jones Beack

Adult recoveries

Dams Bonneville trap HcNary trap

1 5

Sport fishery I)-arm Lake Deschutes River

Indian ceremonial fishery

0 0

r-l

Hatchertes Klickt tat SFH Leavenworth NFH

Tot al 52

9,2&l 11,326 16,239 241 344 876

31 31 85

2 0 0

0 0

0

0 6

7

134 91

1 n 0 0

1 n

1 0

1 n

0 1 9 n

4 1

al - Recoveries adjusted for sampling effort.

b/ - Includes two fish observed previously at the HcNarp trap.

48

delayed mortality as described by Park et al. (1981). After 14 days, survival

in the samples averaged 94% (range 90-99%) Newcomb4/).

Adult returns from experimental groups were not consistent with the

relative outmigrant suvival indicated by juvenile sampling. Equal outmigrant

survivals for spring and fall-marked control releases were not reflected in

adult returns. Instead, 48 fish from the fall-marked release returned to

Leavenworth NFH, but only four fish returned from the spring marked release.

Similarly, very low adult returns were observed for the groups transported to

White Bluffs and Dalton Point (Table a).

Drastically reduced survival was common to all groups handled in the

spring. Although ultimate survival was affected, it was not due to short-term

mortality from stress of handling or transportation as indicated by the high

rate of recovery of juveniles at dams and at Jones Beach. Instead, spring

handling apparently predisposed these fish to extreme mortality later, perhaps

following ocean entry. A causative fact could be related to disease. Results

of physiology studies conducted during the spring of 1980 (Volume III, Novotny

and Zaugg 1984) confirmed the presence of BKD organisms in 80 and 66% of the

spring chinook salmon sampled on 31 March and 28 April, respectively. Studies

holding spring chinook salmon in seawater for 3 months lend credence to this

hypothesis Bjornn-5/). Groups of these fish experienced high losses in

seawater due to BKD. Some groups of fish which showed an incidence of BKD as

low as 1% (as determined by Indirect Fluorescent Antibody Tests) when

-------4 /- Dr. Timothy Newcomb, Natl. Mar. Fish. Serv., 2725 Montlake Blvd. E.,Seattle, Washington 96112, pers. comm. 1986.

5/- Dr. T. C. Bjornn, Idaho Cooperative Fishery research Unit, University ofIdaho, Moscow, Idaho, pers. comm. 1986.

49

introduced to seawater, showed an increasing incidence up to 70% and a severe

loss when held in seawater.

If we assume from the sampling at Jones Beach that spring chinook salmon

juveniles were surviving to the lower Columbia River, then the principal

period of high losses occurred from the Columbia River estuary to the time

these fish would have grown to a size that could legally be retained in the

various ocean fisheries. The poor survival of fish held in the seawater plus

the latent mortality from the heavy infestations of BKD strongly indicates

that the period of greatest mortality occurred from entry into seawater

through the first winter at sea.

Decreased adult returns were also evident for the fall-marked control

group, although not to the extent seen for experimental groups handled in the

spring. All the 1982 and 1983 returns at Leavenworth NFH were subject to

biological sampling according to procedures established by USFWS. Results

indicate a return of approximately 2,900 fish 0.203%) from 1,423,000 unmarked

spring chinook salmon released in 1980. The percentage return from the fall-

marked control 0.050%) was significantly less (p < 0.01, df = 1). Handling

and marking may have also influenced survival of this group, even though the

fish were marked in November and not subjected to further manipulation.

Kooskia NFH, 1980

Background

This experiment (conducted by I C F R U ) was concerned with using a

relatively short distance migration to imprint juvenile spring chinook salmon

to be transported and released at a downstream site to return as adults to

their hatchery of origin; Juveniles reared at Kooskia NFH were allowed to

migrate luu m in a flume, collected, transported by truck, and released into

50

the Columbia River below Bonneville Dam. Data from these fish were compared

to data from fish released from the hatchery to migrate downstream

naturally. The experiment had the following objectives:

1. Determine the effectiveness of the short distance migration

imprinting technique in returning adults to the hatchery in comparison to the

normal hatchery release method (natural migration imprinting).

2. Determine the effect of the two types of imprinting techniques on

overall survival.

Experimental Design

Both the normal-migration and migration-transport groups of spring

chinook salmon released from Kooskia NFH in 1980 were tagged (CUT) and fin

clipped before any migration was allowed. The normal-migration group was

flushed from the raceways and out of the hatchery on 16 April 1980. The

migration-transport group was then allowed to migrate voluntarily out of the

raceways and across the hatchery in the effluent flume (approximately 100 m)

before they were trapped, placed in a truck, and transported to Lower Granite

Dam and then to the Lower Columbia River (Bjornn and Ringe 1984). The

voluntary migration of the migration-transport group took place over a U-day

period (23 April to 5 Nay). Fish used in the 1980 releases were yearling

smolts that averaged 131 mm total length when released.

Results

Adult recoveries were insufficient for analysis [see Bjornn and Ringe

(1984) for summary of adult recoveries].

51

Rapid River S F H 1980

Background

This experiment (conducted by ICFRU) also used a short distance migration

to imprint juvenile spring chinook salmon to be transported and released at a

downstream site to return as adults to their hatchery of origin. Juveniles

reared at Rapid River SFH were allowed to migrate 4 km, then collected from

Rapid River, transported by truck, and released into the Columbia River below

Bonneville Dam. Data from this test were compared to data from fish released

from the hatchery to migrate downstream naturally. The experiment had the

following objectives:

1. Determine the effectiveness of the short distance migration

imprinting technique in returning adults to the hatchery in comparison to the

normal hatchery release method (natural migration imprinting).

2. Determine the effect of the two types of imprinting techniques on

overall survival.

Experimental Design

A group of fish marked in November 1979 by Idaho Department of Fish and

Game (IDFG) personnel for a contribution to fisheries study was used as the

normal MIGRARTION group (control) from Rapid River SFH. Fish were taken from a

rearing pond, tagged with a CUT, fin clipped, branded, and then released into

the hatchery effluent channel. The channel was not screened so the fish could

leave and migrate downstream during the winter or early spring if they chose

to do so. Voluntary migration out of the rearing ponds during the fall and

winter is normally allowed at Rapid River SFH.

52

The limited migration group (test) was allowed a migration of 4 km, and

then they were collected from Rapid River,, marked, and transported by truck in

April for release downstream from Bonneville Dam (Bjornn and Ringe 1984).

Results

Adult recoveries from the experiment were insufficient for statistical

analysis. A total of 27 marked adults was recovered at the Rapid River SFH--

25 from the control releases and 2 from the transported test group [see Bjornn

and Ringe (1984), for Summary of adult recoveries].

Overview of Imprint Treatments for Spring Chinook Salmon

Research objectives relative to the spring chinook salmon homing

experiments were not realized because adult recoveries from all experimental

groups were too small to provide a meaningful statistical analysis.

53

FALL CHINOOK SALMON EXPERIMENTS

Fall chinook salmon were used in four homing experiments (Table 1).

These fish represented two discrete genotypes commonly designated "tule" and

"upriver bright" chinook salmon. The designation "tule" is applied to fall

chinook salmon that are in an advanced state of maturation when they return as

adults to the Columbia River. Tule stock chinook salmon used in our mid and

lower river experiments were from Spring Creek N F H and Big Creek SFd (ODFW)

(Fig. 7). The fish were progeny of naturally returning brood stock adapted to

hatchery culture in the mid and lower Columbia River areas.

Upriver bright chinook salmon return to the Columbia River as green fish

(not sexually mature), still retaining their silvery ocean coloration. The

stock of fish we used migrates over 470 miles to reach their spawning grounds

in the Snake River. Our test fish were from brood stock trapped at Ice harbor

Dam on the Snake River and reared at the Hagerman NFH in Idaho.

Spring Creek NFH - White Salmon - Stavebolt Creek, 1979

Background

This experiment was concerned with imprinting fall chinook salmon reared

at a mid-river site to home as adults to a lower river homing site. Juveniles

from Spring Creek N F H reared for a limited time in the White Salmon River

Rearing Channel were subjected to natural migration and single imprinting at

the Stavebolt Creek Homing Site. Data from these lower river releases were

compared to data from releases of naturally migrating fish from the 'White

Salmon River Rearing Channel. With some modifications, this study is a

replicate of the 1978 Willard-Stavebolt Creek experiment with coho salmon.

This study was designed to assess:

54

Figure 7 .--Study area germane to homing experiments with fallchinook salmon.

55

1. The level of smoltification at which juvenile fall chinook salmon are

most receptive to a homing cue.

2. The time period required to imprint a homing cue.

3. The effectiveness of single and natural migration imprinting in

enabling adults to return to a specific homing site below their hatchery of

origin.

Experimental Design

Eleven groups totaling 554,586 fish were marked at Spring Creek NFH and

moved to the White Salmon River Rearing Channel (a satellite rearing facility)

where discrete holding was possible (Fig. 7). Fish were held and reared for 9

to 44 days before release. Test groups were transported by truck to the

homing site on Stavebolt Creek. Following holding periods of 4 and 48 h, the

fish were released into Stavebolt Creek or into the Columbia River at Hammond,

Oregon. Control groups were released into the White Salmon River (additional

detail on group treatment may be found in Volume II, Table C1.0)

The first release series was made between 28 and 31 March, the second

between 17 and 22 May, and the third on 26 June 1979. Additional details of

the experimental design were given in Slatick et al. (1980). Random samples

from the study population of fall chinook salmon were sacrificed to determine

physiological condition and health. Live samples were transported to

Manchester, Washington, and held in the marine net-pens for observation of

seawater adaptation. Additional details on methods used to measure fish

condition and health are contained in Volume III, Novotny and Zaugg 1981.

56


The three series of test releases were scheduled to coincide with rising,

peak, and declining NA+-K+ ATPase enzyme levels (Fig. 8). Timing of the

releases was based on the 1978 NA+-K+ ATPase activity profile for fall chinook

salmon at the Spring Creek NFH. The plan to release fish at three clearly

different levels of enzyme activity was not executed for three reasons: (1)

peak NA+-K+ ATPase activity in 1979 was about 10 days earlier than in 1978,

(2) fall chinook salmon held at the White Salmon Rearing Channel never reached

the expected level of enzyme activity, and (3) the third release series was

delayed by an outbreak of Enteric Redmouth Disease.

The disease outbreak reduced the number of fish available for the third

release and forced a change in the experimental design. Even though treatment

was applied, 32% of the original marked group died. By the time treatment was

completed and the disease was controlled, high water temperatures in Stavebolt

Creek prevented the planned transfer of test groups; therefore all fish from

the third series were released as controls into the White Salmon River.

At the time of release, most of the fish in the third release series were

clinically healthy (Volume III, Novotny and Zaugg 1981). However, adult

recoveries from these mid-river control releases were extremely low--only four

tags (0.004%) were recovered compared to 141 fish f r o mn the second release of

controls and 207 from the first release (Table 9). The reason for the severe

losses in the third control release is unknown; it may have been because of

the late June release period, or more likely, it was due to latent effects of

the epizootic disease outbreak.

57

20

15

>.Z.gzi

0$ 10a

‘u+4

‘r

Third

Firstrelease

Secondrelease

originalscheduledrelease

I’ iI’ iii

IIII

II IIII

I II I IfIfII’

Ii

IIII ; I; I IfIf 11

II II II II,II, tt II II II II ,! f,! f IIIIIIII III I II

yiiII

1313 2020 2727 33 1010 1717 2424 11 88 1515 2222 2929 55 1212 1919 2626

Thirddelayedscheduledrelease

Ii

II

f f

II

I!

III I

III I

-- March -- -----April-------- --May--- ----June----

Figure 8. --Gill Na+-K+ ATPase activity profiles for fall chinooksalmon at Spring Creek NFH 1978 and 1979 and Big WhiteSalmon River Rearing Channel 1979.

58

Table 9.--ReccwerIes of tags from control and test groups of fal I chinook salmon taken in the ocean andColumbia River flshertes, hatcheries, and on the spwntng grounds AS Juveniles, these f Ish wereheld In the White Salmon River Rearing Channels and then transpotied and lmprlnted to StaveboltCreek for 4- and 48-h periods and released In two locations. Recoveries were from September 1980to December 1982.

JuvenI let released Mul t racoverlesColun&la Total

Experlmentalgroups ND.a / Date

Oceam(no-1

River(no.)

recoveryNo. I

T/C?rat10

First release series (28 to 31 March 1979)

Control(White SalmonRtver release)

Natural lmprl nt(Stavebolt release)

48 h

Single Imprint(Hannond release)

48 h

ControlWhite SalmonRiver release)

Natural Imprint(Stavebol t release)

4&40 h

S1 ngle lmpr Int(lhmond release)

4&48 h

controlWhite SalmonRiver release) 99,669

Total 470,027

42,419

44,337

44,401

47,788

93,821

95,592

28 Mar 109 98 207

30 Mar 205 185 390

31 Har 165 165 330

Second release series (17 to 22 Hay 1979)

19 May 77 64 141

17 619 Hay 62 44 106

22 May 13 8 21

Third release series (26 June 1979)

26 Jun 3 1 4

634 565 1,199

0.488

0.880 1 .80: 1**

0.743 1.52:1**

.0295

0.111 0.38:1

0.022 0 . 0 7 : 1

0.004

a/5

AdJusted for lnftlal tag loss.- Test/control ratlo 1s based on total reccwerl es.* = Slgnl fkart dl fference betueen test and control group (P < 0.01, df = 1).

59

A test of the hypothesis that fall chinook salmon released at different

levels of NatK+ATPase activity may show different degrees of homing requires

significant adult returns to the homing site. Because of the disease problem

in the third release and the reduced survival of the second release, the

number of adult returns were insufficient to determine the relationship

between gill NatK+ ATPase activity and homing of fall chinook salmon. Rata

from adults recovered from the first and second releases though, did provide

other useful information on homing and survival of various treatment groups.

Recoveries of marked salmon from stream surveys were insufficient to

determine differences between 4- and 48-h imprint times but did show the

treatment provided a positive homing response to the Stavebolt Creek area. No

marked fish were actually recovered in Stavebolt Creek, but 20 marked fish

were recovered in the Lewis and Clark River inwithin 4 miles of the creek. No

marked fish were recovered in the other three river systems draining into

Youngs Bay that contained spawning fall chinook salmon. The lack of spawnfng

in Stavebolt Creek might have been due to rejection of the creek by adults

because of extremely low water flows in the creek at the time of spawning.

Data from stream surveys indicated, as they did for coho salmon, that

homing of fish released in Stavebolt Creek (natural imprint) was better than

for those released at Hammond (single imprint) after being imprinted in

Stavebolt Creek. Ten of the fourteen fish recovered from the March release

and five of the six fish from the May release were Stavebolt Creek releases

(Volume II, Tables Cl.1 and cl.11).

Comparisons between recoveries in the various fisheries, hatcheries, and

spawning grounds best illustrate the positive homing response of both the

60

Stavebolt Creek and Hammond releases to Youngs Bay and vicinity (Table 10).

Approximately 60% of all test fish recovered from the 1st release series were

recovered in Youngs bay-of the 350 total recoveries from the first series,

all but 16 were recovered below the Cowlitz River, and only 4 were recovered

above Bonneville Dam. Data from the second release were similar, but totals

were much lower.

The influence of the homing imprint on the control and test lots is

reflected in the two major areas Where the returning adults were recovered.

Test fish, which were imprinted to Stavebolt Creek, were recovered principally

in the Youngs Bay and lower Columnia River area (97 to 98%) whereas 90 to 94%

of the adults from the control release were recovered in the Bonneville area

(Table 10).

Survival of fish released in March was almost five times higher than

survival of those released in May. A factor which may have influenced the

lower survival of fall chinook salmon from the second release series may have

been the latent effects of pathogenic infection combined with stress induced

by handling and transport during the experimental releases. Organ tissue

taken on 19 May from fish held for the second release series indicated

exposure to some type of pathogenic infection--probably ERM and/or B K D (Volume

III, Novotny and Zaugg 1 9 8 1 The White Salmon River group (least stressed)

produced 2.6 times more adults than those transported and released at

Stavebolt Creek (U.295 VS 0 011% return), and those released in Stavebolt

Creek produced five times as many adults as those transported to Stavebolt

Creek, held, and transported again to Hammond for release. The differences

were significant (P < 0.01, df = 1). In contrast, transported fish from the

61

Table 10.--A canparlson betveen recovery peroemtages of the various treatment groups of adult fal I chinook salmonat vsrlous location in the Columbia River. As juveniles these flsh were held in the White SalmonRearing Channels and then transported and imprinted to stavebolt c r e e k for 4- to 48-h periods andreleased in two locations

HomingImprint

Releaselocation

No. ofadultsrecovered

Percentages of adults recovaed a /

white TotalYoungs zone l-5 Zone 6 Salmon Below AboveBay f lshery fishery River Strays Bonneville Bonnevlll

1st Release Series

Natural(Control 1

WhiteSalmon R.

Natural Stavebolt(Test) 46 h Ckeek

Single Hanmond(Test) 46 h

2nd Release Series

98

185

165

Natural White(Control 1 Salmon R. 64

Natural Stavebolt(Test)4 &48 h Creek 44

0 9 23 0 67 10 90

59 29 1 0 11 97 3

59 29 0 0 12 97 3

0 6 39 5 50 6

50 34 0 0 16 98

94

2

Slngle Hammond(Test)

4 & 48 h &’ 75 12 0 0 12 100 0

a/- Numbers rounded off to neaest percent.

b/- Very few fish represented In percentages shown.

62

March release of apparently healthy fish produced over 1.6 times as many adult

fish as the control releases (significant at p < 0.01, df = 1). These data

suggest a strong correlation between the degree of handling and stress and the

latent effects of pathogenic infections on survival of chinook salmon.

The data also suggest that the pathogenic infections encountered in this

experiment did not appear to affect the imprinting of these fall chinook

salmon. Although, the survival of fish in the second lower river release

series was drastically reduced, the riverine migratory behavior of the adults

was similar to fish from the first lower river release (Table 10).

These data strongly suggest that if fish are transported to a lower river

homing site, contributions to ocean and bower Columbia River fisheries May be

enhanced if the release is made when the juveniles will accept and retain a

homing imprint. In contrast to the 1978 coho salmon experiments, fall chinook

salmon did not home to the Stavebolt Creek homing site, but some fish did home

to the Lewis and Clark River in the vicinity of Stavebolt Creek.

Conclusions

1. Homing of fish released in stavebolt Creek was better than homing of

those released at Hammond after being imprinted in stavebolt Creek.

2. Test fish generally homed to Youngs Bay and contributed principally

to the Zone 1-5 fisheries. Control fish returned to areas above Bonneville

Dam and contributed principally to the Zone 6 fishery.

3. Recoveries of fish released in March were six times higher than for

those released in May. Survival of those released in June was nil.

4. Transported fish from the March release produced over 1.6 times as

many adult fish as the control releases, and there was no signif icant

difference between the rate of return of fish released in Stavebolt Creek and

/ those released in brackish water at Hammond.

63

3. Data obtained suggest a positive correlation between the degree of

handling and stress and the latent effect of pathogenic infections on survival

of chinook salmon.

6. The locations of adult riverine recoveries imply that the pathogenic

infections encountered did not appear to affect the imprinting of these fall

chinook salmon.

64

Background

Big Creek S F H - Stavebolt Creek, 1980

The goal of this experiment was to imprint juvenile fall chinook salmon

from a lower river hatchery to a lower river homing site located in a nearby

drainage system. Juveniles from the Big Creek SFH )ODFW) at Knappa, Oregon,

were subjected to natural migration and single imprinting in combination with

exposure to a limited short distance migration in Stavebolt Creek (Fig. 7).

The results of these imprint tests were compared to those for fish which

migrated naturally from Big Creek SFH. The objectives of the experiment were

to:

1. Determine if the riverine adult migratory behavior of the production

release from Big Creek SFH was the same as the subpopulation of fish used in

our experiments.

2. Determine the relative effectiveness of the two imprint techniques in

returning adults to the Youngs Bay drainage system and the Stavebolt Creek

honing site.

3. Compare overall survival and riverine adult migratory behavior of

test fish imprinted to Stavebolt Creek to fish that migrated from the Big

Creek SFH.

Experimental Design

The experimental design consisted of a control group and two test groups

utilizing juvenile fall chinook salmon from the Big Creek SFH. Groups of

112,000 to 15,000 unmarked juveniles were hauled 30 miles by truck daily from

Big Creek SFH to the homing site on Stavebolt Creek over an d-day period.

After a short migration o f 600 feet, the fish were recaptured and marked.

Fish in Test Group 1 (49,528 fish) received 4 to 6 h of exposure to Stavebolt

65

Creek water and then were transported to the West Mooring Basin at Astoria,

Oregon, and released into the Columbia River immediately above the confluence

with Youngs Bay-single imprint. Fish in Test Group 2 (50,414 fish) received

6 to 9 h exposure to stavebolt creek water before being released into

Stavebolt Creek-natural imprint. The control group of 43,863 fish was marked

and released at Big Creek NFH.

A group of 142,400 juveniles was also marked from a random sample of the

entire hatchery production as part of the fall chinook salmon hatchery

evaluation study. These fish were premarked by Oregon Department of Fish and

Wildlife personnel and released 13 May 1980. This marked production release

enabled us to compare the behavior of the subpopulation of fish used in our

experiment to the behavior of the total salmon population reared and released

at the Big Creek SFH (see Volume II, Table C2.2 for additional detail on

experimental design).


A comparison of adult recoveries from our experimental control release

and the hatchery evaluation release showed a close similarity in their

migratory behavior. These data are based on a sample of the population which

returned to the Columbia River. There were no significant differences between

the proportions of these two groups of adults recovered in the Zone 1 gill-net

fishery, returning to the Big Creek environs, or straying to other tributary

systems IN the lower Columbia River (Fig. 9). These data demonstrate that the

behavior of fish from the subpopulation used in our experiment was

representative of the Big Creek SFH fall chinook salmon population, and that

differences in behavior by fish in the test groups would be the result of

behavior modification induced by the experimental treatments.

66

80 - BIG CREEK HATCHERY

60

20

0n=

Fishery Big Creekarea

I 0 Hatchery evaluatione Experimntal control I

55 9

Strays

Figure 9.--Comparison of tag recovery locations of adult fall chinooksalmon in the Columbia River system from two marked groupsof juveniles released at the Big Creek SFH in 1980.Recoveries are through December 1983.

67

There were significant differences in homing between fish from the

control release and the two experimental treatments. There were also

significant differences in homing between the two experimental treatments. As

expected, the majority of adults from the control release homed to Big

Creek. A total of 61% of the recoveries were in the Big Creek homing area;

this included the Big Creek terminal fishery, spawning fish in Big Creek, and

the Big Creek SFH (Table 11). Twenty-one percent of the fish strayed to other

tributaries within a radius of 24 miles, one fish (2%) was recovered from the

gill net fishery in Youngs Bay, and six fish (14%) were recovered in the

Zone 1 fishery (Table 11, Fig. 10). Additional detail on returns from

specific releases are in Volume 11, Tables C2.1 to C2.4.

Adults from the stavebolt Creek release demonstrated a strong positive

homing response to Youngs Bay. A total of 29 recoveries (65%) were in the

Youngs Bay area and only 1 recovery in the Big Creek area. Of this number,

four fish (9%) were recovered in the Lewis and Clark River within 4 miles of

the imprint site in Stavebolt Creek (Table 11). No marked fish were recovered

in the other two river systems that drain into Youngs Bay and contained

spawning fall chinook salmon. The remaining 14 recoveries (31%) were from the

Zone 1 fishery adjacent to Youngs Bay. This indicates a positive response for

homing to the Stavebolt Creek area (Slatick et al. 1984).

Adults from the Astoria test release did not show as positive a homing

response to the Youngs Bay area as fish from the Stavebolt Creek release.

Only 36% of the fish released at Astoria homed to Youngs Bay-signicantly

less than the 65% return from the Stavebolt Creek release (P < 0.05, df = 1)

(Table 11). One fish (2%) was recovered in the Lewis and Clark River and none

in the Stavebolt Creek trap. Numbers of recoveries in the Zone 1 fishery were

comparable to those from the Stavebolt Creek release.

68

Table ll.-A comparison between recoveries in various fisheries and spawning escapementlocations in the Columbia River of adult fall chinook salmon from the 1980 Big CreekStavebolt Creek experiment. Recoveries are through December 1983.

Adults recovered at variousfocations in

Columbia RiveraControl Test 1 Test 2

Revovery Big Creek Astoria Stavebolt Creekarea release release release

X No. X NO. X No.

Commercial fisheries

Zone 1Youngs BayBig Creek

Subtotal

14.0 62.0 12.0

18.018

26.0 1734.0 222.0 1

62.0 40+*

31.056.00.0

87.0

1425

039 *?

Spawning escapement

Lewis and Clark River 0.0 0 2.0 1 9.0 4Big Creek Hatchery 52.0 22 22.0 14 2-O 1Big Creek 7.0 3 2.0 1 0.0 0Other tributaries 21.0 9 14.0 9 2.0 1

Subtotal 80.0 34 38.0 2 5 13.0 6

Total adults recoveredin Columbia River 42 65 NS 45 NS

- Numbers rounded off to nearest percent.a/

!!I Recovery locations include Bear Creek, Gnat Creek, and Plympton Creek in Oregon and GraysRiver, Skamokawa Creek, Elokoman River, and Abernathy Creek in Washington.

** - Significant difference between test and control releases (P < 0.01, df - 1).'NS = Nonsignificant

69

G52EL 40

r88f 20

.

n= 36 52 53Oceanfishery

8 4039

ColumbiaRiverfishery

34 25 6Spawningescapement

-p<O.Ol.df=lIndicates significant difference betweentest and control group

Figure 10. --A comparison of the distribution of adult recoveriesfrom control and test releases of juveniles in the1980 Big Creek-Stavebolt Creek experiments. Recoveriesare through December 1983.

70

Many of the fish from the Astoria release that did not home to the Youngs

Bay area or were not captured in the lower river fisheries continued their

migration up the Columbia River to the Big Creek area (hatchery of origin).

The percentage return of these fish to the Big Creek SFH was 64% of the

returns from the control release.

Total tag recoveries frow both the ocean and Columbia River indicate that

fish from the Astoria test release had an enhanced relative survival over

those released as controls at the hatchery (T/C ratio of l-33:1). Recoveries

from the Stavebolt Creek test release showed a 1.09:1 T/C ratio; however,

neither increase was statistically significant (Fig. 10). Although there were

no significant overall differences between test and control recoveries back to

the Columbia River, there were signif icant dif fereuces between test and

control release vith respect to contributions to the riverine commercial

fisheries and spawning escapement. Both test groups contributed significantly

(P < 0.01, df = 1) more fish to the fishery whereas significantly (P < 0.01,

df = 1) more control than test f fish were from the spawning escapement

recovered in the Big Creek environs and other Columbia River tributaries (Fig.

10).

Adults returning from the Astoria release were harvested at a rate

equally as high as the Stavebolt Creek release in all areas (over three times

greater than the control release) and returned to the hatchery in numbers

equalling 64% of the control release. The rate of return to Big Creek SFH

from the 1980 release was 0.1%; a return of 0.067 would be more than

sufficient for an egg take.

Conclusions

1. The behavior of fish from the subpopulation used in our experiment

was representative of the Big Creek SFH fall chinook salmon population, any

71

differences in behavior by fish from the test groups were the result of

behavior modification induced by the experimental treatments.

2. Adults from the Stavebolt Creek release demonstrated a positive

homing response to Youngs Bay.

3. Adults from the Astoria test release did not show as positive a

homing response to Youngs Bay as did fish from the Stavebolt Creek release.

Most of those that did not home to the bay homed back to Big Creek. Numbers

returning to the hatchery were 64% of the return of the control groups.

I4. The modified (altered) migratory behavior of adults induced by the

experimental treatments affected the numbers of fish entering the spawning

escapement or harvested in the river fishery. Test releases contributed

significantly more fish to the fisheries whereas control fish contributed

significantly more fish to the spawning escapement.

5. Adults returning from the Astoria release had an equally high rate of

harvest as the Stavebolt Creek release in the river fishery (over three times

greater than the control release) and a return to the hatchery of 64% of the

control releases.

6. Imprint and release techniques like those used in the Astoria release

could provide significantly more fall chinook salmon to the Columbia River

fishery than releasing juveniles directly from the hatchery while providing

adequate returns to the hatchery for egg take if the rates of return were

similar to those measured for the 1980 releases.

Spring Creek NFH, 1980

Background

This experiment was concerned with imprinting juvenile fall chinook

salmon from a mid-river site that were transported and released below

72

Bonneville Dam to return as adults to their hatchery of origin. Juveniles

reared at Spring Creek NFH were subjected to sequential imprint techniques

that included being transported by barge before being released into the

Columbia River below Bonneville Dam (Fig. 11). The results of these test

releases were compared to those of fish released from Spring Creek NFH to

migrate downstream naturally. The objectives of the experiment were to:

1. Determine the relative effectiveness of the sequential imprinting

techniques in returning adults to the hatchery of origin.

2. Determine the effect of the sequential imprinting techniques on

overal 1 surviva 1.

For the evaluation of this experiment, our sampling locations were

grouped into two areas: (1) the “Bonneville area,” from Tanner Creek upriver

to Hood River and (2) the “below-Bonneville area,” the Columbia River in Zone

l-5 (Fig. 11).

Experimental Design

The experimental &sign consisted of a control group and two test groups

utilizing 259,786 marked fall chinook salmon from Spring Creek NFH. One

experimental group of 99,583 fish was pumped directly from the raceways into a

barge ; the second group of 99,703 fish was crowded through a 350-ft transport

channel before being pumped into the barge. Both groups were given sequential

homing cues by being transported to a release site below Bonneville Dam by a

barge initially containing Spring Creek water and then Columbia River water.

The control group of 60,500 fish was marked by USFWS personnel as part of the

fall chinook salmon hatchery evaluation study. Additional details of the

experimental design are given in Volume 11, Table C3.0.

w .I.*- . ..I.----.WA3HINti I UN

I.

I- Zones l-5 ,P>I+Zone5 fisbry1 (extends to McNary Dam,+

I km 470)

YLittle Whit:

f

9

/

l yhd~erSalmon i?.

t

%-

9-l _

\ /

k:zrkx

Homing site-Spring Creek WuI-. . .l Other salmon hatcheries

II

-- Barge route to release site IL

Figure 11 .--Location map of release sites and recovery areas for 1980Spring Creek homing study.

74


This experiment may have been impacted by the eruption of Mount St.

Helens on 18 May 1980. Juveniles in the control group were released from

Spring Creek NFH on 9 May and migrated seaward under normal river

conditions. Median passage of this group at the Jones Beach sampling site was

12-14 May (Dawley et al. 1981). Fish for the two test groups were loaded into

the barge and released below Bonneville Dam on 19 May, 1 day after the

volcanic eruption. During their seaward migration, the test fish had to

contend with the plume of volcanic debris emitting from the Cowlitz River.

Median passage of the test fish at Jones Beach was 25 May. There is evidence

from Dawley et al. (1981) that survival of subyearling chinook salmon was

adversely impacted by the eruption.

Differences in rate of adult recovery of test and control fish in the

fishery and back at the Spring Creek NFH homing site indicated that the

techniques used to implant a homing imprint in the juvenfle fall chinook

salmon were not completely successful. Up to two-thirds as many adults from

the barged groups returned to Spring Creek NFH as did adults from the control

release (Table 12). The lower recovery rates for fish from the test lots were

statistically significant from the recovery rates for the control lot at the

hatchery (P < 0.01, df = 1).

A large number of adults strayed to other hatcheries in the Bonneville

area (Volume 11, Tables C3.1 to C3.3). Straying was more prevalent for fish

from the test groups than from the control group. Of the total hatchery

recoveries, about 73% of the test fish and 14% of the control fish were

recovered as strays to other hatcheries, primarily the Bonneville Hatchery

(Table 13). The straying rate of control fish (14%) indicated that a 100%

Table 12.~-Percentage return of fall I chinook salmon at hatcheries and In ocean and Colubmia River flshewles that were released as controlor test groups of smolts following lmprlntlng to the Sprlng creek NFH in 1980 Recoveries are through December 1983

Percentge return of fall Chinook salmonBonneville area

Experimentalgroup

Juveniles released Bonneville TotalSprlng Creek area hatchery Below Bonneville area

NO. Data homing site hatcheries recovery Zone 6 Total Zone 1-3 Ocean Total Total

Control 60,500 09 May 0.200 0.033 0.253 0.200 0.433 0.095 0.388 0.483 0.916(Spring Creekrelease)

Test 1(Loaded racewayand barged)

99,585 19 May . 0 1 3 4

Test 2 99,703 19 May 0 . 1 0 4(Loaded channeland barged)

Total 259,786

0.390** 0.524*, 0.076** 0.600", 0.101 NS 0.410 NS 0.511 NS 1.111 l *

0.267** 0.371,' 0.081** 0.452 NS 0.093 NS 0.345 NS .0438 NS 0.890 NS

NS = Nonsignificant difference between test and control groups.** - Significant difference between test and control groups (P < 0.001, df I 1).

76

Table 13.-A comparison of hatchery recoveries at the homing site and as straysto other hatcheries for fall chinook and coho salmon from the 1980Spring Creek and Willard NFH homing experiment.

Experimentalgroups

1980 Spring Creek fall chinook salmon

Adult recoveries-at hatcherieshomings i t e Other hatcheries%: No. % No.

Control 86.0 121 14.0 20Barge Test 1 26.0 133 74.0 388Barge Test 2 28.0 104 72.0 265

1980 Willard Coho Salmon

Control 98.0 252 2.0 4Combined barge test 89 .0 201 11.0 25

imprinting rate may not be feasible with this stock of fish. The 74 and 72%

straying rates infer that a large proportion of the juveniles (from Test Lots

1 and 2, respectively) did not imprint a homing cue when they were loaded into

the barge containing Spring Greek NFH water. We believe that the short period

these juveniles were in Spring Greek NFH water in the barge (20 min and 1 h 55

min for Test Lots 1 and 2, respectively) was insufficient for the majority of

the fish to receive a positive homing imprint.

As previously stated, coho salmon juveniles which had been held in a

barge containing Little White Salmon River water for 19 to 21 h exhibited a

strong positive homing imprint (see also Slatick et al. 1982). Of the total

hatchery recoveries of adult coho salmon, 89% of the fish from the barged test

groups and 98% of the fish from the control group returned to the Little White

Salmon NFH homing site (Table 13). It is very possible that a longer imprint

time (approximately 24 h) inn a barge containing Spring Greek NFH water would

also give a more positive homing cue to fall chinook salmon smolts to return

as adults to the Spring Greek NFH homing site.

The data indicate that even though outmigrants from the barged test lots

were released into the Columbia River the day after Mount St. Helens erupted

(19 May), their relative survival equalled or surpassed that of the control

release that migrated downriver prior to the eruption. Fish from Test Group 1

had a significantly higher overall recovery rate than did fish from the

control release (Table 12) (P < 0.01, df = 1). Survival of Test Group 2 was

much lower and similar to survival of fish from the control release. The

stress induced by the extra handling that juveniles received when they were

crowded through the transport channel before being pumped into the barge may

have contributed to the lower survival rate of the second test group.

78

There were significant differences in recoveries of fish from the test

and control lots by various user groups in the Columbia giver system. up to

twice as many barged as control fish were recovered at hatcheries in the

Bonneville area. Because of lack of imprinting, significantly more fish from

barged groups than from the control group were recovered in hatcheries other

than the Spring Creek NF H homing site (P < 0 . 0 1 , df = 1). Conversely,

significantly more fish from the control group than from the barged groups

were recovered at the Spring Creek NFH and also in the Zone 6 fishery

(Table 12) (P < 0.01, df = 21111 Recoveries in the ocean and Zone 1-5 fishery

area showed no significant difference in the numbers of fish taken from either

the barged or control lots.

Conclusions

1. Methods used to implant a honing cue in test groups of- juvenile fall

chinook salmon barged below Bonneville Da m were only partially successful, the

barged fish returned up to two-thirds as many adults to spring Creek NFH as

the control release.

2. About 14% of the control lot strayed; therefore, 100% imprinting

probably cannot be achieved with this stock at Spring Creek NFH.

3. Barged fish from Test Group 1 produced significantly more adult fish

than the control release; the majority of these adults were recovered in the

Bonneville area.

79

Background

Hagerman NFH 1980

This experiment was concerned with imprinting juvenile fall chinook

salmon to be transported and released into the Colmbia giver below Bonneville

Dam to return as adults to the Snake River. Juveniles reared at Hagerman NFH

in I&ho were subjected to a natural migration imprint and a single imprint in

combination with a limited short distance migration before being transported

by truck and released into the Colmbia RIver below Bonneville Dam. The

objectives of the experiment were:

1. Determine the relative effectiveness of the two imprinting techniques

in returning adults to the Snake River.

2. Define the effect of the two imprinting techniques on overall

survival.

Experimental Design

Fish used in this experiment were part of the Snake River fall chinook

salmon egg bank program. Adults were collected in September 1979 at Ice

harbor Dam and held at the Tucannon SFH [Washington Department of Game

(WDG) 1 l Eyed eggs were then shipped to Hagerman N F H where the fish were

reared until they appeared to be smolts. The fish were tagged (CUT) and had

their adipose fin excised in May 1980 and transported from the hatchery in

early June.

T h e normal migration group was transported by truck and released in the

Snake River near Asotin, Washington, on 3 June 1980. The migration-transport

group (limited migration) was transported by truck to Lower Granite Dam on 5

June 1980. and released into a raceway. The fish then received a limited

migration imprint by being allowed to voluntarily move out of a raceway

80

(Bjornn and Ringe 1984). The fish were then trapped and transported by truck

to the lower Columbia River.


No fish with CWTs were sacrificed at Lower Granite Dam when the fall

chinook salmon were moving downstream in 1980. However , most , if not all, the

fish with adipose clips that entered the collection facility during June and

early July were probably fall chinook salmon released at Asotin, Washington.

NMFS personnel estimated, on the basis of adipose-clipped fish collected

during June, that 3,425 of the 60,750 fall chinook salmon released at Asotin

were collected at Lower Granite Dam.

Fall chinook salmon that migrated out of the raceway at Lower Granite Dam

and then were transported by truck to the lower Columbia River were recaptured

in signif icant ly larger numbers in the estuary sampling than those released

at Asotin (P < 0.01, df = 1). Only 13 of the Asotin-released fish were

collected in the estuary samples vs 46 of the migration-transport fish

(Table 14).

Adult fall chinook salmon from the group released at Asotin (normal

migration) returned to the Snake River at 31 times the rate of adults from the

group released in the raceway at Lower Granite Dam and transported to the

lower Columbia River (Table 14). Reported recaptures of the normal-migration

group through December 1983 were relatively high (0.38% overall) with 57 fish

recaptured in ocean fisheries, 5 in Columbia River fisheries, and 170 at Ice

Harbor and Lower Granite Dams. In contrast, only 28 (0.05%) of the transport

group were recaptured, 20 in the ocean fishery, 3 in the river fishery, and 5

at Snake River dams. Based on ocean recoveries, survival of fish from the

Table 14.-Fall chinook salmon smolts released in the Snake River in 1980 andadults recaptured for the migration-homing study.

__ -.--Normal-migration Migration-transport

Item group group

Number of fish marked and releasedwith coded wire tags 60,750 57,713

Wire tag code 5/527 5/518

Date released 3 June 80 6-23 June 80

Mean total length at release (mm)(n=E9)

Smolts recaptured in the estuar& 13 4 6

Adults recaptured (through July 83)

Ocean fisheries 57 2 0

Columbia River 5 3

Snake River dams 170 5

Adults recaptured %

At Snake River dams 0.280 0.009

Total 0.382 0.049- - - - --- --

a/ Fish with coded wire tags.z = Significant difference between test and control groups (P<0.Ol,df = 1).

82

migration-transport group was significantly less than fish from the normal

migration (control) group P < 0.01, df = 1).

Conclusions

1. The limited migration method used combined with truck transport was

unsuccessful in returning adult fall chinook salmon to the Snake River.

Except for ocean recoveries, returns were insufficient for analysis.

2. Based on ocean recoveries, survival of fish from the migration

transport group was significantly less then for fish from the normal migration

group (P < 0.01, df = 1).

83

Overview of Imprint Treatments of Fall Chinook Salmon

Recoveries of adult fall chinook salmon returning to the Colmbia River

system indicated that the combination of imprint method, mode of

transportation, release site, timing, and physiological condition of the

juveniles influenced their future adult migratory behavior pattern.

Natural Migration Imprint

The majority of adults from the hatchery of origin releases returned to

their respective homing sites at Big Creek and Spring Creek hatcheries or were

recovered in nearby gill-net fisheries. Straying by adults occurred from both

releases.

Although juvenile salmon released at the lower river site (Stavebolt

Creek) were from two different stocks [Big Creek SFH in 1980 and White Salmon

river gearing Channels in 1979 (Spring Creek NFH stock)], the returning adults

behaved in a similar wanner. The adults demonstrated a positive homing

response to the Youngs Bay drainage area.

Adults returning from the mid-river area imprinting site (White Salmon

River Rearing Channels) demonstrated a positive response to the Bonneville

area. Although these juvenile salmon were held and reared from 9 to 44 days

before release, homing of adults back to the Unite Salmon RIver was poor. The

majority of these fall chinook salmon strayed to hatcheries in the area, with

most returning to the Spring Creek NFH--their hatchery of origin.

The straying behavior of adult fall chinook salmou from these lower and

mid-river release sites corresponded very closely to conclusions on straying

drawn by Lister et al. (1981): (1) the rate of straying increases with

decreasing distance between the release and rearing sites, (2) a relatively

nigh proportion of the straying is back to the rearing site, and (3) straying

84

rates tend to decrease with increasing distance between the rearing and

downstream release sites.

Single Exposure imprint

Adults returning from juveniles subjected to the single exposure imprint

did not show as positive a homing response as did fish from the natural

migration imprint. Fish from the mid-river (Spring Creek stock) which

received a single imprint to Stavebolt Creek and were released at Hammond in

1979 showed a significantly (P < 0.01, df = 1) greater positive homing

response to the Youngs Bay area than fish from the lower river area (Big Creek

stock) which had been imprinted to Stavebolt Creek and released at Astoria in

1980.

Many of the fish from the Astoria release that did not home to the Youngs

Bay area continued their migration up the Columbia River to the Big Creek area

(hatchery of origin). In contrast, very few of the fish released at Hammond

were recovered in areas other than Youngs Bay.

Sequential ixposure Imprint

Adult recoveries at the Spring Creek NFH homing site indicated that

barging to implant a sequential homing imprint in juvenile fall chinook salmon

was partly successful. Up to two-thirds as many adults from the test lots as

from the control returned to the hatchery. This homing response may be

improved by a longer imprint period in the barge. The data indicate that

riverine homing behavior by adult fall chinook and coho salmon in the lower

Columbia River was similar (homing response to Youngs Bay area) when subjected

to the same imprint treatments as juveniles. This suggests that a longer

imprint time may improve homing for fall chinook salmon since it was

successful with coho salmon.

85


1. Fish surplus to an upriver hatchery’s need, as with coho salmon,

could be transported to Youngs Bay, given a single exposure or natural

migration imprint, and released. This should provide more adults by reducing

dam-related mortalities to smolts and provide an area (Youngs Bay) where the

fish could be harvested without impacting other runs of fish.

2. Fish surplus to a lower river hatchery, such as Big Creek SFH, given

a single exposure imprint could provide greater numbers of fish to the

riverine fishery and provide sufficient fish for spawning at the hatchery.

3. Direct barging from Spring Creek NFH to below Bonneville Dam with

proper sequential exposure imprinting could provide an enhanced fishery and

still return sufficient numbers of fish to the hatchery for egg taking.

4. Surplus juvenile fall chinook salmon can also be used to enhance the

river fisheries above Bonneville Dam by imprinting them to pre-selected

harvest areas and barging the smolts to below Bonneville Dam for release.

This would reduce dam-related mortalities to smolts and provide greater adult

returns to the harvest areas.

86

STEELHEAD EXPERIMENTS

Steelhead were used in nine homing experiments (Table 1). These fish

included three upriver racial stocks and one lower river stock. The upriver

stocks normally travel over 450 miles on their spawning migration whereas the

lower river stock normally migrates about 140 miles.

During our studies, steelhead from the Dworshak NFH were the only

indigenous stock of fish which were imprinted to their hatchery of origin.

These fish were originally from the North Fork of the Clearwater River

(Dworshak N F H water supply) and were adapted to hatchery culture. Fish used

for the Chelan-Leavenworth and Wells-Winthrop experiments were from brood

stock taken from fishways at upper mid-Columbia River (above McNary Dam) dams,

and included a combination of both hatchery and wild fish. Brood stock for

the Chelan SFH ( D W G ) experiment were obtained from the fishway at Priest

Rapids Dam, and fish for the Wells SFH (WDG) experiment were obtained from the

Wells Dam fishway. Steelhead used in the Tucannon SFH studies in 1978 and

1979 (Snake River system) were Skamania stock, a lower river race from the

Washougal River, Washington. A map showing the various homing and release

sites, sport fishery, and the Zone 6 Indian fishery is contained in Figure 12.

Background

Dworshak N F H , 1978

Juvenile steelhead reared at Dworshak N F H were subjected to single and

sequential imprinting with hatchery water prior to being transported and

released below Bonneville Dam. Aaults from these releases returning to the

hatchery were compared to adult returns from naturally imprinted fish released

at the hatchery. The objectives of the experienced were to:

87

pizz-x:::sj;: Sport fhing uom

Luvenrrorth NF;I

\ LinNGO068 I

\ ’ 1. - - n

I •:&~lndi8nset~-~jgillnetfisho~ : I

I I1 I

Figure 12. --Study area germane to homing experiments with steelhead.

1. Determine if juveni le steelhead transported and released at

downstream sites can be imprinted to return as adults to their hatchery of

origin.

2. Define the relative effectiveness of three imprinting techniques.

Experimental Design

A control group of CWT steelhead was released at Dworshak NFH into the

North Fork of the Clearwater RIver (hatchery water supply), and two test

groups of CUT steelhead were imprinted with North Fork Clearwater River water,

transported, and released below Bonneville Dam. The test groups were taken

from the recondiiioned hatchery water and held in raw North Fork Clearwater

River water for 6 days. One group was trucked to Lewiston, Idaho; held

overnight; and barged downstream to the Bonneville Dam release site

(sequential imprint). The second test group was trucked directly to

Bonneville Dam and released (single imprit). Additional detail on

experimental design may be found in Volume II, Table D1.O.


Returns of adult steelhead to the Dworshak NFH indicate the test methods

used were successful in varying degrees in returning steelhead to the Dworshak

NFH homing site. Homing of both barged and trucked transport groups was

impaired as indicated by T/C ratios. A T/C ratio of over 3:l was indicated

for transported fish returning to the lower Columbia River compared to 1.54: 1

for barged and 1.08:1 for trucked fish back at the hatchery (Table 15). Even

though homing of both test groups was impaired, sufficient homing cues were

imparted to fish in the barged group to allow a significantly higher return of

barged fish than control fish to the hatchery (P < 0.01, df=l).

89

Table 15.-Returns to flve sampling locations and to the Dworshak homlng site of steelhead from control and testreleases of smolts imprinted to the Dworshak NFH in 1978. recoveries were from September 1979 to 12Nay 1981.

Recovery areaand experimenta/

Bonneville DamC/

Juveniles releasedND. Date ND.

Adult returnsr T/C ratlo

Dworshak - controlTrucked to Bonnevl I leBarged to Bonneville

100 60&20:661

21 Apr 13 0.04301 MAY 16 0.324 7.53:l NS

24,006 36 Apr 9 0.157 3.65:1 NS

Indian flshery ed

Oworshak - controlTrucked to 8onnevll IeBarged to Bonnavfl Ie

FcNary Dam=’Dworshak - controlTrucked to Bonnevl I leBarged to 8onnevll le

Lover eanlte Dard

Duorshsk - controlTrucked to 8onnev 1 I qBarged to BonnevIl le

75 0.075U 0.21361 0.254

21 0.0704 0.0889 0.158

198 0.65819 0.37350 0.932

Clearwater and Snake Rlver sport e/fisheryDvorshak - control 76 0.076Trucked to 8onnevllleBarged to 8onnavll le

Drorshak blng s i t e

8 0.03922 0.154

2.84~1 -3.39:l l *

1.2&l l

2.26:l l

0.57:l l *1.42:l *+

0.51:l NS2.02:l l *

Dwrshak - amtrol 280 0.278Trucked to 8onnevlIle 62 0.300 1.08:l NSBarged to Bannevllle 103 0.429 1.54:l l *

“/ Because of dlfferences In sampling lntenslty (efffclency) at each trapping slle, results are not conparablebatueem SItas.

? Adjusted for the differences In detectabillty between binary end color-coded wire tags as IndIcsted byreturns to Dworshak hatchery.

r/ Data from branded flsh only.%! A total of 100,600 were wire tagged for the hatchery control releases. Of thls number only 30,074 were

branded for inriver adult evaluation.E! Data from coded wire tags only.

NS = Nonslgnlflcantl = Slgnlflcant difference betveen the test and control group (P c 0.05, df = 1).

l * = Slgnlflcant difference between the test and contol group (P c 0.01, df I 1).

90

Effects of transportation on the relative survival and homing of the test

groups trucked or barged were demonstrated by recoveries in the two principal

fisheries (Zone 6 Indian fishery and Clearwater River harvest) and returns to

the Dworshak NFH homing site. The total estimated (minimum) recovery of 2-

ocean age adults was 1.44% for the trucked fish and 1.97% for the barged fish;

both were significantly higher than the estimated 0.92% recovery for the

control fish (P < 0.05, df = 1) ( T a b l e 16). These figures reflect the

increased survival and subsequent contribution to user groups of the test lots

transported directly fron the Dworshak NFH compared to the higher losses from

the control lot (not transported from Dworshak NFH). The difference in rate

of return of test and control fish is even more impressive when one considers

that approximately 67%̂ of the control fish surviving to Lower Granite Dam were

also transported below Bonneville Dam via the regular transportation program,

thus providing them greater survival potential (Park et al. 1979). The T/C

ratio for returning adults from the 1978 outmigration transported from the

collector dams was 3.22:1. This means that approximately three out of four

returning adult steelhead from our control release received the benefit of

being transported around hydroelectric dams on the Snake and Columbia

rivers. It is apparent that without the benefit from transportation, the

return of control fish to the hatchery would have been substantially less.

The impaired homing resulted in a large number of the test fish delaying

or remaining in the Bonneville Pool as evidenced by the catches in the Indian

fishery (Table 16). Nearly 90% of the control fish were taken in the fall

fishery during the upstream migration. In contrast, nearly 75% of the test

fish taken were those that overwintered in the Bonneville Pool and were caught

in the winter gill-net fishery.

91

Table 16.-Minimum estimated recovery of 2-ocean age steelhead in Indianfishery (Zone 6), Clearwater River harvest, and actual recoveries atDworshak NFH homing site from control and test releases of smoltsimprinted to the Dworshak NFA in 1978.

Location and-period ofrecovery a/

(2::Xt)blNo. x No. x No. x--

Indian fisher&

FallWinterSubtotal

Clearwater Riverharvest

190 21 7124 113 113214 6.213 134 0.649 184 0.766*

467 0.464 lU7 0.518 NS 188 0.783”

Dworshak NFH(homing site) 249 0 .248 57 0.276 NS 100 0.417*

Total 930 0.924 298 1.442* 472 1.966*

a/- Because of differences in recovery (efficiency) at each location, resultsare not comparable between sites.

b! Number of juveniles released.

Cl- Estimated recoveries based on sampling of Zone 6 Indian fishery.

d/- Estimated recovery of both Indian and sport fisheries based on totalestimated Clearwater River harvest by Idaho Fish and Game--(Pettit 198r;e/).

NS = Nonsignificant.

* = Significant differences between the test and control group (P < 6.05,df = 1).

e/ Steve Pettit, Biologist, Idaho Department of Fish and Game, 1540 Warner,Lewiston, ID 83501, pers. commun. 1982.

92

A key point is that even though homing of the barged group was impaired,

there were still enough fish that received a positive homing imprint to

provide a significantly greater percentage return of adults to the hatchery

and to the Clearwater River sport fishery than did fish released at the

hatchery (P < 0.01, df=1).

Conclusions

1. Steelhead can be imprinted to return to their hatchery of origin

after being transported directly from the hatchery as smolts and released

below Bonneville Dam.

2. Steelhead smolts sequentially imprinted by barging returned more than

1.5 times as Many adults to the hatchery as smolts that received a single

imprint and were trucked directly to below Bonneville Dam, or those released

at the hatchery.

3. With the river conditions impacting the outmigration of steelhead in

the Columbia River system, returns from the barged group provided the first

evidence that fish imprinted and transported directly from a hatchery will

return as adults in greater numbers than fish that migrated naturally

(controls).

4. Survival of barged and trucked fish was significantly higher than

survival of fish released at the hatchery. The difference in rate of return

would have been even greater had survival of 67% of the control releases not

been enhanced by also being transported around the hydroelectric dams on the

Snake and Columbia Rivers.

5. Test fish that did not imprint to their hatchery of origin returned

to the area near where they were released as juveniles.

93

Background

DWORSHAK NFH, 1980

This experiment was concerned with using a relatively short distance

migration to imprint juvenile steelhead to be transported and released at

another site to return as adults to their hatchery of origin. Juveniles

reared at Dworshak NFH and allowed to volitionally migrate from their rearing

area were trapped within the hatchery system, transported, and released into

the Columbia River below Bonneville Dam. Adult returns to the hatchery from

the test group were compared to adult returns of fish released from the

hatchery to migrate naturally. The objectives of the experiment were to:

1. Determine if this imprint technique would return adults back to the

hatchery as well or better than the normal hatchery release method (natural

migration imprinting).

2. Determine the effect of these two imprinting techniques on overall

survival of the test groups.

Experimental Design

Age-i steelhead produced at Dworshak NFH were used for this experiment.

The control group was tagged by IDFG personnel as part of their hatchery

contribution studies. On 17 April 1980, 59,125 fish were tagged with CWTs and

then released by flushing the holding ponds into the mainstem Clearwater

River.

The test group was tagged after the fish voluntarily migrated out of the

hatchery ponds down an effluent sluiceway and into a trap. Trapping and

marking started on 28 April and finished on 30 April. During the 3 days,

40,010 migrants were trapped and tagged and 8,490 of the tagged fish were also

branded (Table 17). Marked fish were hauled to Lower Granite Dam 29 April

94

Table 17.-Steelhead trout smolts released from Dworshak NFH in 1980 and adultsrecaptured for the migration-homing study.

--Normal-migration Migration-transport

Item Group group

Number of fish marked and released

Coded wire tags

Brands

Wire tag code

Brand used

Date released

Wean total length at release (mm)

Smolts recaptured in the estuary

Adults recaptured

Ocean fisheries

Deschutes River

Columbia River sport & net fisheries

Others

Idaho fishery

Dworshak NFH

Adults recaptured (%)

In Idaho

Total

59,125

5/4/55

-

17 April 80

185

106

0 1

0 4

61 2 2 4

6 3

37 8

152 7 1

0.320

0.433

40,010

8,490

10/21/19

LD 4(4)

29 April to2 May80

199

160**

0.197**

0.777**

a/ Based on recovery of CWT fish (Bjornn and Ringe, 1984).** - P < 0.01, DF - 1; indicates significant difference between test and

control groups.

95

through 2 May and transferred to barges or trucks for transport to the lower

Columbia River (Bjornn and ringe 1984)


Timing of the smolt migration through the estuary was spread through

5 weeks (24 April to 2 June) for control fish and 1 week (3 to 9 May) for test

fish. At the estuary (Jones Beach), NMFS personnel (Dayley et al. 1981

collected significantly fewer marked steelhead from the control group then

from the test group (p < 6.61, df = i) (Table 17). Probably more test fish

reached the estuary than control fish because of enhanced survival from

transportion. It is also likely that fewer test fish were in a non-smolting

condition at the time of release since these fish were all voluntary migrants

from the hatchery ponds. The control group included all fish in the ponds and

could have included unsmolted fish that did not migrate downriver

Homing of test fish was impaired as indicated by the significantly

greater rate of return from the control group to the Clearwater River

(P < 0.01, df = 1) (Table 17). In contrast, the overall rate of return of the

test group (0.78%) was nearly twice the rate of the control group (0.43%).

This difference was also significant (P < 0.01, df f 1). Many of the test

fish recoveries, though, were from the lower Columbia River fisheries in early

spring--an indication they were lost and milling in the Bonneville pool.

Conclusions

1. The limited migration method used was not completely successful in

returning adult steelhead to the Dworshak NFH homing site; significantly more

fish returned from the control release than from the test release.

96

2. Significantly greater numbers of adults were recovered in the

Columbia River system from the test (transported) releases than from the

control (non-transportedj releases.

3. Many of the test fish (transported) remained in the vicinity of

Bonneville Dam where they were taken in large numbers in the early winter

fishery in Zone 6

97

Tucannon SFH, 1973

Background

This experiment was concerned with imprinting juvenile steelhead to a

unique segment of a hatchery water supply before they were transported and

released at a downriver site to enable adults to return to the hatchery homing

site. Skamania stock (lower river) steelhead juveniles reared at the Tucannon

SFH were subjected to sequential imprinting with hatchery spring water and

migration route waters prior to release into the Columbia River below

Bonneville Dam. The objectives of the experiment were to:

1. Determine the effectiveness of two sequential imprint techniques in

returning adults to the homing site.

2. Determine the effect of sequential imprint techniques on the relative

survival of this stock compared to normal hatchery release procedures.

Experimental Design

The spring water portion of the hatchery water supply was used as the

initial homing cue. Two groups of fish which had been maintained on 100%

Tucannon River water were removed from the hatchery ponds and held 1 h in a

tank truck while the composition of the water supply to the ponds was

altered. The two groups of test fish were then returned to separate ponds,

one of which contained 100% spring water and the other a 20:80% mixture of

spring:Tucannon River water. Following a 48-h holding period, the fish were

transported by truck around the 34 miles of Tucannon River they would have

encountered during a natural outmigration, loaded into a barge moored at Lyons

Ferry on the Snake River (near Little Goose Dam), transported to below

Bonneville Dam (RM 140), and released on 17 May, providing sequential exposure

to the Snake and Columbia river waters along the barge route.

98

A group of the same stock reared at the Tucannon Hatchery was released

into the Grande Ronde River between 30 April and 10 May by the WDG. Relative

survival and homing of tests lots were compared with this group. Additional

detail on experimental design may be found in Volume 11, Table D2.0.


Returns of adults indicate that the imprinting techniques used were

unsuccessful in returning the test groups of steelhead to the Tucannon

Hatchery homing site. No fish were recovered at the hatchery or in our

sampling of the Tucannon Rfver. Imprint methods used, however, did implant

sufficient homing cues to enable as many of the 100% spring water barge group

to return to the Snake River as the Grande Ronde River group (Table 18). In

contrast, significantly less 20% spring water fish than either the 100% spring

water or Grande Ronde group returned to the Snake River (P < 0.05, df = 1).

There was no observed straying of test fish to the Columbia River above its

confluence with the Snake River. By comparison, five of the fish released in

the Grande Ronde River were recovered in the Wenatchee River sport fishery

(Volume II, Tables D2.1 to D2.3). This would indicate that straying can be

caused by other reasons than transportation and lack of imprinting.

Relative survival of the 100% spring water group (based on overall rate

of return) was higher than either the 20% spring water group or the control

release (Table 19). Both test groups provided 11 times as many fish to user

groups (primarily the Zone 6 Indian Fishery) as the control release.

Besides enhanced surviva 1, the main reason for the higher catch rate of

test fish was probably the fact that many of the test fish that had not

accepted a homing inprint as juveniles returned and milled in the vicinity of

their release site and were more susceptible to the fishery.

99

Table 18.-Returns to four sampling locations of steelhead from controland test releases of smolts from the Tucannon SFH in 1978.Recoveries were from June 1979 to 30 November 1981.

Sampling Adult returnslocation NO.

and juvenilesexperiment released NO3 X T/C ratio

Bonneville DamGrande Ronde R., control100% spring water, test20% spring water, test

55,557 24 0.04318,137 54 0.298 6.93:1 *18,549 28 0.151 3.51:1 *

Indian fisheryGrande Ronde R., control100% spring water, test20% spring water, test

8 0.01427 0.149 10.64:1 *29 0.156 11.14:1 *

McNary DamGranle Ronde R., control100% spring water, test20% spring water, test

3 0.0053 0.017 3.4O:l NS6 0.032 6.40:1 NS

110 0.19838 0.201 1.02:l NS11 0.059 0.30:1 *

Lower Granite DamGrande Rode R., control100% spring water, test20% spring water, test

a/- Because of differences in sampling intensity (efficiency) at each trapping site,results are not comparable between sites.

NS = Nonsignificant

* = Signif icant difference between the test and control group (P < 0.05, df - 1).

100

Table lg.--. Yinimum estimated recovery of adult steelhead in the Indian fishery and Lower Granite Dam and actual recoveries in the sport fisheries below Lower Granite Dam from control and test releases of smelts imprinted to the Tucannon SFH in 1978.

lOOX spring water 2Od spriq water Location and period of

a/ recoverv-

Control (55, 557)bJ

No. % No.

barge (18,547)y

No. %

C/ Indian fishery (Zone 6)

Fall 5 Spring 14

Subtotal 19

Sport fish 35 iecd/ Hatcheries

Subtotal

Total 27

e/ Lower Granite Dam-

Fall Spring

8 0

8

442 152 42 1 0 0

Subtotal 443 0.797

Grand total 470 0.846

0.034

0.014

0.049

25 32 52 45 77 0.425 77

29 1

30

107

15 15

0.165 30

0 59& . 107

152

259

0.840 42 0.226

1 42of’ . 149 0 805' .

0.415

0.162

0.577f'

a/ Because of differences in recovery (efficiency) at each location, results are - not comparable between sites.

b/ Nunber of juveniles released. -

C/ Estimated recoveries based on sampling of Zone 6 Indian fishery. -2

d/ Actual recoveries. -

e/ Estimated recoveries are based on recoveries of jaw tagged vs coded wire tagged only adult steelhead at hatcheries upriver from Lower Granite Dam from control and test releases of juveniles from the transportation study in 1978.

f/ Totals for barged fish: 107 + 107 214 259 + 149 = 18,137 + 18,547 = 36,684 = oe5832, 36,684 1.1127 0

101

Conclusions

1. Adults from the test groups failed to return to the Tucannon SFH

homing site.

2. During the barging process, a portion of the 100% spring water test

group received a homing cue which enabled as many adults to home to the Snake

River as adults returning to the Snake River from the control release in the

Grande Ronde River.

3. Those test fish failing to imprint to the Snake River returned as

adults to, and remained in, the Columbia River or its tributaries below the

confluence of the Snake diver.

4. The combination of impaired homing and enhanced survival of

transported fish resulted in barged releases providing over 11 times as many

fish to the sport and Indian fisheries as did control releases.

Experimental Design

Tucannon SFH, 1979

This experiment was a repeat (with slight modification) of the 1978

Tucannon S F H experiment. Th e test fish were held in a tanker for 2 h instead

of 1 n, and the control fish were released into the 'iucannon River instead of

the Grande Ronde River. The release into the Tucannon River reduced the

distance the natural migration group had to migrate in the Snake River by 107

miles and exposed them to two less hydroelectric dams than were experienced by

fish in 1978. However, releasing these fish below the two uppermost dams also

deprived them of the benefits of being included in the ongoing fish

transportation program. T h e release into the Tucannon River modified our

objectives to:

102

1. Determine the relative effectiveness of sequential and natural

imprinting in returning adult steelhead to the Tucannon SFH homing site.

2. Determine the relative effect of three imprinting techniques on

overa 1 surviva 1.


The homing behavior of adult steelhead, transported as smolts in 1979,

was similar to that seen in the 1978 experiment (Slatick et al. 1982). While

no adults from test or control groups returned to the Tucannon Hatchery homing

site, a portion of the test fish received a homing cue which enabled some

adults to return to the Snake River. Because only one control fish was

recovered at Lower Granite Dam, an accurate assessment of homing of test

groups could not be made.

Transporting fish from the hatchery to below Bonneville Dam enhanced

surviva 1. More than 16 times as many of the 100% spring water test group

returned as adults to the Bonneville Dam sampling site as did controls (Table

20). Also, survival of fish from the 100% spring water test group was

signif icant ly higher than survival of fish from the 20X spring water test

group (P < 0.01, df = 1). The 16:l transport benefit was over twice the

6.93: 1 benefit for the 100% spring water test group measured in 1978. The

increased benefit was more likely due to poorer survival of control releases

than to enhanced survival of test fish in 1979. There were only 2 adult

recoveries of control releases at Bonneville Dam and 1 at Lower Granite Dam,

compared to 24 and 110, respectively in 1978. In addition, those controls

that did survive exhibited considerable straying in 1979 (discussed in detail

in Steelhead Overview Section).

103

Table 20.--Complete returns to four sampling locatins of steelhead from control andtest releases of smolts from the Tucannon SFA 1979. Recoveriesfrom June 1980 to November 1983.

Sampling No.and juveniles

experiment released N o .

-Adult return

xb- T/C ratio

Bonneville Dam

Tucannon, control 24,787 2 0.008-100% spring water, test 20,728 27 0.130 ** 16.25:1

20% spring water, test 22,058 7 0.031 3.88:1

Indian fishery

Tucannon, control 0 0.000100% spring water, test 26 0.12520% spring water, test 18 0.081

McNary Dam

Tucannon, control 6 0.024100% spring water, test 0 0.00020% spring water, test 2 0.009 O-38:1

Lower Granite Dam

Tucannon, control 1 0.004100% spring water, test 6 0.028 7.O:l20% spring water, test 1 0.004 l.O:l

a/ Because of differences in sampling intensity (efficiency) at each trapping site,results are not comparable between sites.

b/ Numbers of controls recovered were too small to test for statisticalsignificance bet*ween control and test groups.

** = Significant difference between two test groups (P < 0.01, df ~ 1).

104

The 0.30% estimated recovery rate of test fish released in 1979

(Table 21) was less that one-third that of the 1.12% estimated recovery of the

1978 release (Table 19), indicating a much lower survival of test fish. This

was partly due to lower survival to the river and partly to adverse river

conditions impacting survival and/or homing of returning adults in 1981.

Adult recoveries in the lower river fisheries indicated that survival of the

1978 release was about 2.5 times higher than the survival of the 1979

release. However, adult recoveries at Lower Granite Dam showed that returns

from 1978 test fish releases were more than eight times that of the 1979

release, indicating an additional 70% loss of fish occurred in 1979 between

the lower and upper river. We suspect that adverse river conditions were to

blame for much of this loss. A majority of adults from this stock of

steelhead migrated over Bonneville Dam from June to mid-July 1981 a period of

high spill at mainstem dams. During this time, the presence of gas bubble

disease in adult steelhead was observed at the Bonneville Dam sampling site

(29 June to 6 July 1981). As they migrated upriver, subsequent exposure could

have resulted in mortality to some fish.

In addition, adults which were imprinted and continued their migration to

the Snake River were confronted with high water temperatures (ranging from 70%

to 78%) from 17 July to 15 September (discussed further in Steelhead Overview

Section).

Conclusions

1. Adults from both test and control groups failed to home to the

Tucannon SFH homing site.

2. During the barging process, a portion of the test fish received a

homing cue which enabled some adults to home to the Snake River.

105

Table 21. -Minimum estimated recovery of steelhead in Indian Fishery (Zone 6)and Priest Rapids and Lower Granite Dam sampling sites, and actualrecoveries in the sport fishery and hatcheries from control and testreleases of smolts imprinted to the Tucannon SFH in 1979.

100% springwater group

20% springwater group

Location andperiod ofrecovery a/ No. X No. X No. X--

Indian fishery(Zone 6)

Fall 0 28 22Winter 0 20 7Subtotal 0 0.000 48 0.232 29 0.131

Columbia Riversystem belowSnake River 0 11

Columbia Riversystem aboveSnake River 5 0

Snake River system 0 1Subtotal 5 0.020 12

Total 5 0.020 60

10

10

0.058 ii. 2 8 9 40

0.0500.181

Priest Rapids Dam 8 0 0

Lower Granite Da& 3 23 4-

Subtotal 11 0.044 23 0.111 4 0.018- - -

Grand total 16 0.065 83 0.40rrel 44 O.l90e/

a/ Because-of differences in recovery (efficiency) at each location, results-are not comparable between sites.

b/ Number of juveniles released.c/ Estimated recoveries based on sampling of Zone 6 Indian fishery.d/ Actual recoveries.

e/ Total for barged fish:2& : 22;:58 -a- o*234x, t:,:8t4 0.297%

I/ Estimated recoveries based on UDF sampling at Priest Rapids Dam.g/ Estimated recoveries are based on recoveries of jaw-tagged vs CUT only adult

steelhead at hatcheries upriver from Lower Granite Dam from control and testreleases of juveniles from the transportation study.

106

3. Test fish failing to imprint to the Snake River returned as adults to

and remained in the Columbia River or its tributaries below the confluence of

the Snake River.

4. The combination of impaired homing and enhanced survival of

transported fish resulted in barged releases providing approximately 11 times

as many fish to the sport and Indian fisheries as control releases.

Tucannon SFH-little Goose Dam, 1980

Background

This experiment was conducted to assess the influence of the

smoltification process (as measured by Na- Atpase enzyme activity) on the

ability of juvenile steelhead to accept a homing imprint. Chelan stock

steelhead reared at the Tucannon SFH were used in the experiment. The

objectives were to:

1. Determine at what phase of the smoltification process juvenile

steelhead are most receptive to imprinting a homing cue.

2. Assess the influence of the Na-+K+ AtPase level at time of release on

the relative survival of Tucannon SFH steelhead.

Experimental Design

NA+-K+ Atpase enzyme activity was monitored at the Tucannon Hatchery from

7 March to 12 June 1980. Figure 13 shows the average profile of Na+-K+ AtPase

activity in relation to dates of release for the three test groups and the WDG

release into the Walla Walla River. The test fish were imprinted with Snake

River water in a raceway at Little Goose Dam for 8 h then transported by truck

and released into the Columbia River below Bonneville Dam. Additional detail

on experimental design may be found in Volume II, Table D4.0.

107

-ii.-i

Fii;$a>.Z.siti

f

:

TEST SERIES

1st 2nd 3rd15 -

10 -Transported

5 -

0 I 1 I I, I 1

b

Walla Walla Riverrelease

8 I I Ip-March,k 4 810 April 18 ,.,.’ 16 30 6 12

May ,-+ June 1

Number of adults recovered 36 110 1 1

Average mm at release 168.6 173.9 135.7 172.8

Figure 13.--Composite Na+-k ATPase profile for steelhead smolts rearedat the Tucannon Hatchery, indicating size at release, numberof adult recoveries, and time frame for imprinting tests in1980. Serial releases of marked transported fish were madeon 8 April, 8 May, and 12 June 1980.

108


Recoveries of adult steelhead in the Snake River system indicated that

juveniles released at or near the peak of the Na- K + ATPase activity profile

(second release) homed back to the Snake River as adults in greater numbers

than adults from juveniles released on the rise (first release) or decline

(third release) of the profile curve (Table 22). However, the actual return

from the 2nd test release was only seven fish 0.035%). This is in contrast

to a recovery of 273 fish 1.591%) from a similar experiment in 1975 using the

same stock of fish (Slatick et al. 1981b). Obviously, the imprinting

technique used in 1980 did not provide the cues needed to return adult fish to

the Snake River Over 8 0 % of the estimated return faiied to imprint to the

Snake River (57 in Snake River vs 274 overall recovery).

The complete lack of adult recoveries from the third Na+-K+ AtPase test

group in the fisheries or at the sampling sites in the mid-Columbia and Snake

Rivers indicated that these juveniles may had reverted to parr and May have

been physiologically unable to imprint a homing cue to the Snake River before

they were transported and released below Bonneville Dam. By 12 June, all size

groups of fish in the third Na+-K+ AtPase release had entered a post-smolt

condition (Volume III, Novotny and Zaugg 1984).

Survival of fish from the second Na+--r+ AtPase release was significantlyL\

greater than from the first release (P < 0.01, dgf = ij. Estimated recoveries

indicated that the second release provided 4.1 times more fish to the Indian

fishery and 1.75 times more fish to the sport fisheries and hatcheries than

the first release (Table 22). Survival of the second release, though, was

much less than fish from the 1978 experiment (i.llX vs 4.14% return [(Table 1)

(Slatick et al. 1 9 8 1 b ) .

Tab lo 22.--Estimatedd return of adult steelhead In vsrlous fisherieis and back at the Lower Granlte Dam sampling site, from releases ofJuveniles impritned to the Walla Walla and Snake Rivers In 1980. recoveries were fromq June 1981 to November 1983.

Locat Ionand

recovery~

Walla Walla r. b/natural migration

17-18 Mayfl7,92#

NO. %

Mo. and % of adults recaptured1st ATPase release

transported?8 Aprl IZ!f21,652jp/

NO. % No.

3rd ATPace releasetranspotied?

12 Jun-C/(18,964)y

NO. I

lndlan f Irhery”Fal IWinter

Subtotal

0 50 37?i 0.000 4 2 0.194

25, 0134 0i!F 0.805 0 0.000

Sport fisheries and hatcheries f/

Columbia Riversystem belowSnake River 0

Columbia Riversystem aboweSnake River 0

7 8

0 0Snake Ri ver system 0

Subtotal Ti 0.0000

7 0.032-

5z 0.056

Total 0 o.ooo 49 0.226 170 0.861

’Lower Granite Da& 16 0.069 0.018 2Grand Total 16 0.069 55 0.245 220 1.114 1 0.005

0

10i 0.005

1 0.005

0 0.000

a/ Because of differences In recovery (efficiency) at each location, results are not comparable between sites.b/ Type of release.c/ Release date.d/ Number of Juveniles released.e/ Estimated recoveries based on sampling the Zone 6 fishery.f / Acutal recoveries.g/ Estimated recoveries are based on recoveries of jaw-tagged vs coded wire-tagged only adult steelhead at hatcharies upriver fromLower Granite Dam from control and test releases of juveniles from the Transportation Program.

110

Conclusions

1. Within the Na+-k+ Atpase levels tested, it appears that releasing

fish near the peak of their na+-k+ Atpase level will promote higher adult

returns.

2. Migratory survival of steelhead juveniles that have not smoltedq or

have reverted to parr (as indicated by na+-k+ AtPase enzyme activity) is very

poor.

3. Compared to an earlier study in 1976, the optimum release strategy

for imprinting a homing cue to the Snake River in juveniles was not achieved

in the 1980 experiment:

Background

Wells SFH - Winthrop NFH, 1978-79

This experiment was concerned with imprinting juvenile steelhead reared

at an upper mid-Columbia River hatchery to home as adults to a hatchery

located further upstream in a tributary river system. Juveniles reared at the

Wells Hatchery were subjected to single, sequential, and natural migration

imprinting at the Winthrop NFH located on the Methow River. Results from

these treatment groups were compared to fish released at the hatchery

production release site in the lower Methow River. The objectives of the

experiment were to:

1. Determine the relative effectiveness of four imprint techniques in

returning adults to the Methow River and the Winthrop N F H homing site.

2. Define the effect of the four imprinting techniques on overall

survival.

111

Experimental Design

In 1978, the experimental design used five groups of steelhead (20,000

fish per group)--a control group held 2 d at Winthrop NFH prior to release at

the hatchery, the production release made directly into the Methow River 0.25

mile upstream from the mouth, and three transport groups. Transport groups

were held 2 to 8 D at the Hatchery in an attempt to imprint them to the

hatchery water prior to transporting them downriver by barge or truck. One

group was trucked in raceway water to a release site below Bonneville Dam; the

second was trucked in raceway water to a barge at Richland, Washington, and

barged downstream to below Bonneville Dam; the third group was trucked in

raceway water and released at tingold Washington.

The test was repeated in 1979, and the experimental design was identical

to the 1978 test except there was no ringold release in 1979 and there were

only 10,000 fish in the group trucked to Bonneville Dam. Additional details

on experimental design may be found in Volume 11, Table D5.0.


Recoveries of steelhead indicate that the imprint methods used were

unsuccessful in returning the test groups of steelhead to the Winthrop NFH

homing site. Only one fish from each test group returned to the hatchery in

1978 and none in 1979. Although imprint methods used in these experiments

were not successful in returning fish to the homing site, they did implant a

limited hodng cue which enabled significantly more fish from all transported

groups than fish from control releases from both the 1978 and 1979 releases to

home to areas above McNary Dam (p < 0.01, df = 1) (Table 23). Homing above

this point was more impaired, as indicated by a decline in1 T/C ratios in both

years for the transport groups at Priest Rapids Dam. The difference in T/C

112

Table 23.--Retllrn.s to five sampling; locations of ndult: steelhead from control and test releases of smoltn from the Wells Hatchery which were imprinted to the Winthrop NFH homlnp, site and the Methow River in 1978 and 1979. Recoveries were from June 1979 to 30 November 1982.

Sampling;

locatlon

Adult returns from 1978 release 20. v I. 1/c rntio

Adult ,returns from 197? release .+I !io. z.l IO f/C ratio

%nn.?t:i.lla 3nn

Cinthrcp XPH 5 0.325 L. Methow River 12 0.060 Truck to Bonneville 36 0.188 Barge to Bonneville 26 0.130 Truck to Ringold 29 0.164

Indian fishery

Winthrop NFII 6 0.030 L. Methow River 14 0.070 Truck to Yonnevi.lIe 49 0.256 Rarge to Bonneville 31 0.155 Truck to Ringold 16 0.091

Mc?Iary Dam

Winthrop NFH 21 0.103 L. Methow River 32 0.161 Truck to Bonneville 90 0.470 Barge to Bonneville 57 0.286 Truck to Ringold 66 0.374

Priest Rapids Dam

Winthrop NFH 38 0.187 L. Methow River 55 0.276 Truck to Bonneville 42 0,220 Barge to Bonneville 23 O;llS Truck to Ringold 81 0.459

Wfnthrop Hominq Stte

Winthrop NFH 19 0.093 L. Methow River 1 0.005 Truck to Bonneville 1 0.005 Barge to Bonneville 1 0.005 Truck to Ringold 1 0.006

2.4:1 NS 7.5:1 * 5,2:1 * 6,6:,1 * ’

2.3:1 NS 8.5:! * 5.2:1 * 3.0: 1 *

1.6:1 NS> 4.6:1 * 2.8:1 * 3,6:1 *

1.5:l NS 1.2:1 NS 0.6:1 * 2.5: 1 *

0,dS:l NS 0.05:1 NS 0.05:1 NS 0.06:1 N!3

8 4

31 35 C/

2

h4

90 Cl

5

0

9

33 11 18 15 2’

0.044 0.020 0.318 0.204

0.011 0.005 0.452 0.525

0.027 0.000 0.092 0.192

0.180 0.055 0.185 0.087

0.022 0.000 0.000 0.000

0,; 5: 1 7.23: 1 ** 4.54:1 **

3.41:1 * 7,ll:l **

0.31: 1 1.03:1 NS 0.48:l *

a/ Because of differences in sampling intensity (efficiency) at each trapping site, Results are not comparable between sites. 11 WDG production release. (S&nple sizes were too small for use in the statistical analysis.) ~1 No truck to Ringold in 1979,

NS = Nonsignificant. * = Significant difference between test and coilrrol groups (P < 0.05, df = 1).

** s Significant difference between test and control groups (P < 0.01, df = 1).

113

ratios at Priest Rapids Dam reflects the varying degrees of homing cues that

resulted from each treatmnt. For returns from the 1978 release, the test

group trucked to Ringold, Washington (2.5:1) was highest, followed by the

group trucked to Bonneville Dam (1.2:1), and the group barged from Richard

Washington (0.6: 1) (Table 23. ). In 1979, there was no Ringold release. The

test group trucked to Bonneville Dam was highest (1.03:1) and returned at

about the same rate as controls. The group barged from Richland, Washington,

was lowest (0.48:1) and as in 1978 returned significantly less fish than the

control group (p < 0.05, df = 1) (Table 23).

Recoveries of tagged fish in the sport fishery (Tables 24 and 25) and at

Lower Granite Dam (Volume II, Tables D5.1 to D5.9) on the Snake River provided

additional data on homing of the various test groups to areas above McNary

Dam. The intensive sport fish sampling by WDG was terminated prior to the

return of 2-ocean age fish from 1979, so for comparative purposes, thei r

estimated contributions to the 1979 sport fisheries contain only l-ocean age

steelhead. Both l- and L-ocean age recoveries were obtained from the 1978

release. The major sport fisheries between McNary Dam and Priest Rapids Dam

are at Ringold, Washington, and a stretch of several miles immediately below

Priest Rapids Dam. The major sport fishing areas above Priest Rapids Dam are

in the Wenatchee, Washington, area and at the mouths of the Entiat and Methow

rivers (Fig. 12).

Data obtained from the 1978 release in the sport fishery generally

verified the data obtained from sampling at Priest Rapids Dam (Table 24). Of

the total sport catch above Priest Rapids Dam, 50% (94 fish) were Ringold

releases, 41% (78 fish) were trucked fish released at Bonnevi lle Dam, and only

9% (18 fish) were the fish barged from Richland to Bonneville Dam. More than

twice as many of the Ringold group, compared to the other two transport

Table 24.--Estlmsted recovery In the sport and Zone 6 Indian fisheries of adult steelhead raturnlng from controland test releases of Juveniles from the 1978 Wells-Wlnthrop experiment. Recoveries ware from 1979 toMarch 1981.

-..

Adult recoveriessport flshervq'

Control Lower Entlat b -,or

testGroups

Juveniles released Columbia Ringold Wonatchee Methow tndlanl'river area area area Total. fishery total

No9 Date (no.1 (no.) (no.) (no.1 (no.) (no.1 No. $ --' T/C ratlo- -.-m

wlnthrq NFH(control)

Lower Methow River(prod. ret. site)

Truck to 8onnevllla(test)

Barge to Bonnevltlet e s t

Truck to Rlngold(test)

Tots I

20,530 06 May 0 0 0 54 54 20 74 0.364

19,901 27 Apr 0 17 14 136 167 47 214 1.075 2.95: t

19,lJl 05 May 0 158 60 18 2 % 155 591 2.044 5.62: t

19,979 04 May 14 103 13 5 135 100 233 1.176 3.25;1

17,637 05 May 14 52 33 41 160 53 213 1,208 3.32:t- - - -

96,978 28 330 140 254 752 375 1,127 1.162

A/- From Hlsata et al. 1979-1980, and Schuck et al. 1980-1981.

b/- AdJusted for tnltlal tag loss.

C/- Estlmeted recoveries based on ranrpllng etflcleney of the Zone 6 lndlan fishery.

115

groups, were able to home to the Methow River as evidenced by the Methoow River

sport catch (41 fish from Ringold Washington, vs 18 from those trucked to

Bonneville Dam and only 5 from those barged to Bonneville Dam).

Because of only 1 year of sampling, returns from the 1979 release were

far less, but the data obtained were comparable (Table 25). Overall, the data

from sampling the sport fishery in both years generally indicated that: (10

barged fish that received an imprint homed to the proximity of the barge

loading area rather than to areas upstream from Priest Rapids Dam-much larger

catches were observed in the Ringold sport fishery than in the Wenatchee-

Entiat sport fishery ’ (Table 24) ; (2) more trucked than barged fish were

imprinted to areas upstream from Priest Rapids Dam, and Bonneville Dam trucked

fish provided more fish to the Ringold sport fishery than barged fish in 1978

(Tables 24 and 25); and (3) Bonneville Dam trucked fish were the only group

with significant straying inio the Snake River (Volume II, Tables D5.1 to

D5.9).

A portion of the juveniles in both the trucked and barged groups in 1979

were apparently not ready or able to accept a homing cue at the time they were

transported. These adults returned to the area near their point of release

near Bonneville Dam and remained there over winter instead of continuing their

migration upriver. As a result, they were more vulnerable to the Indian gill

net fishery for a longer period of time. This was demonstrated by the

recovery from the 1979 release of 100 test fish and no control fish in the

winter fishery compared to 34 test and 3 control fish in the fall fishery

(Volume 11, Tables D5.6 to D5.9).

Transporting fish around dams in 1978 and 1979 significantly enhanced

surviva 1, especially the trucked groups (P < 0.01, df = 1). Between 7.2 and

7.5 times as many trucked fish and 4.6 to 5.2 times as many barged fish

Table 25.--Estimated recoveries at four sampling locations ol l-ocean age adult steelhead from control andtest releases of juveniles from the 1979 Wells-Winthrop experiment. Recoveries were f rom June1 9 8 0 to March 1981

Controlor

t e s tgroups

ISport PislierPJ

Entiat &

Adult recover&s

b/Ri ngold Wenatchee Methow I n d i a n

Juveniles released area area area Total f Q shery TotalNo. Date (no.) (no.) ( 110 . ) ( n o . ) (no. ) n o . % T/C r a t i o

Winthrop NFH(control) 18,298

Methow R.( p r o d . rel. s i t e ) 2 0 , 0 5 2

Truck toBonnevi 1 le( t e s t ) 9,741

Barge to le

09 May

14 May

05 May

0 0 34 34 0 34 0.186

0 I) 20 20 0 20 0.100 0.54:l

18 10 6 34 43 77 0.790 4.25: 1=:a-

17,152 25 Apr 96 19 - 0 -.- 2 0 9 1.219 6.55:l

Total 65,243 54 29 60 143 197 340 0.521

a / F r o m Schuck e t a l . 1980-1981.

h/ A d j u s t e d f o r Initial tag l o s s .

c/ Estimated recoveries based on sampling efficiency of the Zone 6 Indian fishery.

117

returned to Bonneville Dam as did fish from control releases (Table 23). The

result was a much greater contribution of transported fish to various sport

and Indian fisheries. In 1978, the overall rate of return to the user groups

was 1.5% for transported fish vs 0.72% for those released in the Methow River

(Table 24). In 1379, the return of transported fish was 1.0%, but the return

of controls w a s only 0.14% (Table 25). The lower rate of return for controls

probably indicated lower survival of control fish in 197Y. The overall

decline in numbers recovered from the 1979 release (340 vs 1,127 from the 1978

release) is probably a combination of less sampling in the sport fishery (1-

ocean only on 1979 releases) and the lower survival of the control release in

1979.

Although not statistically significant, the recovery of adults from the

Winthrop NFH control release in 1978 was consistently lower than the Lower

M e t h o w River production release at all four inriver sampling sites (Table

2 3 The lower survival probably resulted from mortality during the juvenile

outmigration in 1978. Sampling of the 1978 smolt outmigration at M c N a r y and

John Day Dams in 1978 shored that the lower Methow River production release

group had a three times greater recovery than the Winthrop NFH control group

at both of these juvenile sampling sites (Sims 6 /

The reverse occurred in the 1979 release; survival of the lower Methow

River production release was less than the survival of the Winthrop N F H

control release at the four inriver sampling sites (Table 23). Again, the

lower survival probably resulted from mortality auring the juvenile

outmigration in 1979 Sampling of the 1979 smolt outmigration at Wanapum,

- ---- -___---------

6/- Carl Sims, N M F S NWAFC, 2725 Montlake Blvd. E., Seattle, WA 9 8 1 1 2 pers.commun. 1 9 8 6

118

Priest Rapids, McNary, and John Day Dams in 1979 showed that the Winthrop N F H

control group had five times greater survival than the lower Methow River

production release group based on the average rate of recovery from these four

juvenile sampling sites (Raymond and Sims 1980).

Conclusions

1. Imprint methods used in 1978 and 1979 were unsuccessful in returning

barged and trucked fish to the homing site but did implant a limited homing

cue which enabled Many of these fish to return to areas above McNary Dam.

2. Transporting of the test fish around dams by truck or barge.

significantly enhanced adult survival hack to the Columbia River.

3. The limited homing imprint and enhanced survival resulted in greater

numbers of transported than control fish being caught by the various user

groups.

4. Natural migration imprinting enables this stock of steelhead to

return as adults to homing sites (i.e., Methow River and Winthrop NFH) located

upstream from their hatchery rearing site.

119

Chelan SFH - Leavenworth NFH, 1978-79

Background

This experiment was concerned with determining the length of time

required to imprint juvenile steelhead reared at an upper mid-Columbia River

hatchery to home as adults to a hatchery located on a tributary river

downstream from the rearing site. Juveniles reared at the Chelan SFH were

subjected to sequential and natural migration imprinting at the Leavenworth

NFH. The Leavenworth NFH is located on the Icicle River, a tributary to the

Wenatchee River, which flows into the Columbia River 51 miles below the Chelan

SFH rearing site. The objectives of the experiment were to:

1. Define the effects of three time periods on imprinting juvenile

steelhead

2. Determine the relative effectiveness of two imprnting techniques in

returning adults to the Wenatchee River and the Leavenworth NFh homing site.

3. Define the effect of the two imprinting techniques on overall

survival.

Experimental Design

For this experiment, three (test and control) groups were trucked to

Leavenworth Hatchery and held in raceways for 10 d, 2 d, or 4 h. T h e t e s t

fish were sequentially imprinted by trucking them in hatchery water (Icicle

River) to a barge at Richland, Washington, and then barging them downriver and

releasing them in the Columbia River below Bonneville Dam. The control fish

were released from the hatchery into the Icicle River (natural migration

imprint). The test was repeated in 1979, and the experimental &sign (by

Larry Brown, WDG) was identical to the prior test except the paired 4-h

treatment of 1978 was changed to 6 h in 1979. Additional detail on

experimental design May be found in Volume II, Table D6.0.

120


The length of time juvenile fish in the control lots were held for

imprinting did not make any difference in the rate of return of adults to the

hatchery. There was no significant difference between the 4-h, 2-d, and 10-d

imprint times in 1978 (g = 0.74, df - 2, p - 0.7) or the 6-h, 2-d, and 10-d

imprint tines in 1979 ( g - 2.37, df - 2, p - 0.3). Similarly, and in most

instances, length of imprint time did not appear to affect rate of return of

test fish to Bonneville Dam, the Zone - Indian fishery, McNary Dam, or Priest

Rapids Dam ) 2 out of 12 samples in 1978 and 1 out of 1 samples in 1979 were

significantly different, p < 0.01, df = 2); therefore, the test groups were

combined to illustrate the trends in migratory homing behavior.

The imprint methods used were unsuccessful in returning transported fish

to the upper river above Priest Rapids Dam and back to the homing site. The

T/C ratios for returning adults to the five sampling locations in 1978

illustrate the increasing loss of homing as these fish moved upstream. The

T/C ratio at Bonneville Dam and the Indian fishery was about 3:1. By the time

the fish reached McNary Dam, the T/C ratio had dropped to 1.4:1. By the time

these adults reached Priest Rapids Dam, the T/C ratio was 0.15:1, indicating

nearly complete impairment of homing (Table 26). In 1979, fewer fish were

imprinted to areas above McNary Dam as indicated by the lower 1.02:1 T/C ratio

at Mc N a r y Dam and the higher 12.94:1 T/C ratio in the Indian fishery (non-

imprinted fish generally return to the Bonneville area, mill for several

months, and are more vulnerable to the Zone 6 fishery).

Table 26.~-Returns to five sampling locations of adult steelhead from paired test and control released ofsmolts imprinted to Leavenworth NFH homing site for varying lengths of time prior torelease in 1978 and 1979. Recoveries were from June 1979 to 30 November 1981.

Experiment 1978 1979and Control - No.of No. of

Sampling orlocation

juveniles No. of % Test to junenlies No. of % Test totest released adults return control ratio released adults return control ratio. . . . - - -

Bonneville Dam, Control 69,863 31 0.044 2.86:1** 67,317 44 0.065Test 68,086 86 0.126 70,500 79 0.112

Indian fishery Control 47 0.067(Zone 6) Test 176 0.258

McNary Control 100 0.143Test 137 0.201

O~rlest Rapids Dam Control 166 0.238Test 24 0.033

Weavenworht NFH Control 64 0.092homing site T e s t 1 0.001

3 . 8 5 : 1 33447

1.41:1*+ 54

0.15:

0.01:

58

1'. 20812

1** 691

0.0490.614

0.0600.082

0.3090.017

0.1030.001

1 . 7 2 : 1

12.94: 1

1.02:lNS

0.06:1'*

O.Ol:lN'

.-

,I/. Rocause of dlfferences In sampllng lntenslty (efflclency) at each trapplng site, results are not comparable botwecns i t e s

M S = Nonslgnlflcant.

ch)c

l.J n Significant difference between the test and control group (P < 0.01, df = 1).

122

The transported fish that received a homing cue that enabled them to

migrate upstream as far as McNary Dam, probably received a homing cue to the

Mid-Colubia River between Richland Washington, and Priest Rapids Dam when

they were loaded and held (approximately 12 to 24 h) in the barge. In both

1978 and 1979: (1) very few adults migrated above Priests Rapids Dam;

(2) between 85 and 90% of the test adults recovered in the sport fishery were

taken in the Ringold area (free flowing portion of the mid-Columbia River

between McNary and Priest Rapids Dams,, Table 27) ; and (3) straying into the

Snake River was minimal [recoveries at Lower Granite Dam consisted of only 25

fish from the barged groups and 1 from the control group (Volume 11,

Tables D6.1 to D6.12).

A portion of the juveniles in the test groups, especially in 1979, were

apparently not ready or able to accept a homing cue at the time they were

transported. These adults returned to the area near their point of release

near Bonneville Dam and remained there over the winter instead of continuing

their migration upriver. This was demonstrated by the recovery of 31 test

fish compared to 1 control fish from the 1978 release and 306 test fish

compared to 5 control fish from the 1979 release in the winter (February)

March) Indian gill-net harvest (Volume II, Tables 06.1 to D6.12). In the

combined fall/winter Indian fishery in both years, the recovery of test fish

was significantly greater than control fish (P < 0.01, df = 1) (Table 27).

Transporting the fish around dams significantly enhanced survival of

adults returning to the Lower Columbia River as indicated by the 2.86;1 T/C

ratio at Bonneville Dam in 1978 and 1.72:1 ratio in 1979 (P < 0.01, df = 10.

Overall recovery of steelhead to the various user groups was 1.35% in 1978 and

1.65% in 1979. Rate of return of test fish was higher than control fish in

123

Table 27.--Estimated recovery of adult steelhead from the 1978-1979 Cheian-Leavenworth experiment insport and Indian fisheries. Recoveries were from June 1979 to March 1982.

contro lor

testgroups

a/sport fisheryLower Ent!at &

Columbia Ringold c/Wenatchee Methow Indian T o t a ll Test to

Rlver area area area fishery recovery contrl(no.) (no. ) (no. ) (no.) Total (no.) No. z r a t i o

1978

Control 2 34 622 29 687 157 844 1.208Test 7 325 48 5 385 580 965 1.417 1.08;1

Total 9 359 670 34 1,072 51 1,809 1.311

1979

control 0 0 960 14 974 63 1,037 1.540Test 0 306 19 0 325 933 1,258 1.784 1.16:l

Total 0 306 979 i i 1,303 996 2,295 1.665

a/- From Hista et al. 1979-1980, and Schuck et al. 1960-1981.

b// Fran observed recoveries--no estimates available.

c/ Estimated recoveries based upon sample efficiency of the Zone 6 lndlan fishery.

124

both years, but the difference was not significant. Control releases

contributed greater numbers of fish to the sport fishery above Priest Rapids

Dam, but test releases contributed greater numbers of fish to the Indian

fishery and the Ringold area sport fishery (Table 27).

Conclusions

1. There was no significant difference between the rate of adult returns

to the homing site in relation to the length of time (4-6 h, 2 d, and 10 d)

the juveniles were held for imprinting.

2. Imprint methods used in 1978-79 in conjunction with truck to barge

transportion of juvenile steelhead were unsuccessful in returning adults to

the homing site.

3. A portion of the fish in the transported treatment groups imprinted

to the mid-Columbia River when they were loaded on to the barge as juveniles.

4. Transporting the juvenile steelhead around dams significantly

enhanced the survival of adults returning to the lower Columbia River.

125

Overview of Imprint Treatments of steelhead

Recoveries of adult steelhead returning to the Columbia River system

indicated that the combination of imprint method, mode of transportation,

release site, timing, and physiological condition of the juveniles influenced

their future adult migratory behavior pattern. The results also suggest that

the genotype of the fish can influence the rate of return of transported fish

to a homing site.

The effects of imprint strategies used in our studies to modify adult

migratory behavior are compared by adjusting the data to illustrte comparable

numbers of adult steelhead recovered from the various sampling locations

(Tables 28, 29, 30). A common release of 100,000 fish per test condition was

used to reflect the numbers of adults which would be recovered.

Natural Imprint

natural migration imprint technique was not uniformly successful in

returning adult steelhead to a hatchery homing site. Indigenous Dworshak

stocks returned the greatest number of adults back to the hatchery homing site

(257 TO 278 Non-indigenous upriver stocks also returned to their

hatchery honing sites but at a significantly lower rate (p < 0.01, df = 1)--60

fish at Winthrop NFH and 85 to 117 fish per group (average 97 fish) at

leavenworth (Table 28). None of the lower river1 stocks imprinted to the

Tucannon SFH returned back to the hatchery homing site as adults.

Migratory behavior of adult returns in the Columbia River system was

quite uniform. Most of the fish were observed in the upper river areas (above

McNary Dam, RM 292) on a direct migration route to their homing sites. Very

few fish strayed to other systems leading away from their migration route.

Adults returning from the 1979 Tucannon SFH release were an exception to this

general migratory pattern. The majority of these fish observed above McNary

Table 28.--Adult racoverles of steelhead trout which as juveniles migrated naturally from seven release locations, (adjusted for arelease of 100,000 fish per test).

Recovery locations

Indigenousupper rlver stocka/

Dworshak DworshakHatchery Hatchery1978 1980

Release locations and year of experimentNonindigenous

Lower rlver stock" Upper river stock Grande Tucannon Wall a Lower W i n t h r o p Leavenwoth-Rone R. Hatchey Walla I?. Methow R.y Hatchery Hatchery1978 1979 1980 1978-1979 1978-1979- 1978-1979

Columbia River to McNary DamBonneville Dam 43 w 45 8 0 40 34 55Zone 6 fishery 73 103 18 0 0 38 21 58Sport fishery 9 0 7 0 0 3 0 4

Sub-total 127 103 70 8 0 81 55 iii

Upper Mid-Columbia and Snake RivresMcNary Dam 70Lower Granite Dam 658priest Radids Dam 0

Sport fishery areasSnake and Clearwater 81Ringold 0Entiat and Wenatchee 0Methow 0

Homing sites 278Other hatcheries 0

Subtotal 1,087Unknown 1

64000

2570

32120

5 24 0 80 67 112198 4 6 0 0 1

0 4 0 165 184 273

4020e/

0209

2

0020

000

5 20 0-

0105

268f/

53 52581 513

053

14260

15

370209723

9021

Total 1,215 424 281 60 6 662 568 1,020

a/ Dworshak stock.b/ Skammanla stock.c/ Chelan stock.d/ Wells stock.e/ homing site evaluation is Snake River data above-Lower Granite Dam )198) and Snake and Clearwater (4).f/ Hcmlng site evaluation is Methow River sport fishery data above (268).g/ Fish In upper mld-Colunbla River are collected for brood stock at Priest Rapids and Wells Dams.

127

Table 29.--Adult recoveries of steelhead trout rhlch as Juveniles recetved a single exposure to a unlacewater source before belng transported by truck to two release locations on the Cot umbla River.( a d j usted for a release of 100,000 f ish per test).

recovery locations

homint sites, yea of experiment and release locationslndlgenous

upper river s t o c kNon-indigenous

Upper river stockSworshak Tucannon HatcheryHatchery Snake River% Winthrop Hatcheryc/

1978 1960 1978 1 9 7 8 979Bonnevil le Bonnevi I le Rlngold Bonnevi Ile

Columbia River to McNary DamsBonnevil le DamZone 6 fisherySport fisheryother

Subtotal

356 16 164 232213 186 102 332

5 21 6 240 3 6 10

574 226 278 598

Upper Hld-Columbla and Snake RlversMcNary DanLower Granite D aRlest RapIds Dam

Sport fishery areasSnake and clearwaterRlngoldEntlat and WenatcheeMethow

Homing sitesOther hatcheries

subtotalUnknown

88 0 374 343425 8 40 298

23 2 459 208

39505

3000

8855

Total 1,464 241 1,649

3000d/

11136102164

679

1 , 5 7 10

0905952

3691,122

0

1,720

a / Dvorshak stock

b/ Chelan stock.

c/ Wel Is stock.

d/ homing site evaluation i s from Snake River data above-Lower granite Dam (6) and Snake and Clearater (3).

Tab le 3 0. --Adult recoveries of steelheadd trout which as Juvenlles received a sequental imprint by being transportedby truck and barge to a release site below Ronenvllle Dam (sdJustod for a release of 100,000 fish per test).

homing alter, and year of experimentlndlgonous _ Non-l adlgenous

Upper river Duorshak Dworshak

Lower rl ver stock Upper rl ver stock

Recovery locat IonsHatchery Hatchery Tucannon Hatchery? Winthrop Hatchery? Leavenwort hY

1978 1980 1978-1979 1976-1979 1978-1979

Columbia River to McNary DamBonnovl I le DamZone 6 flsherySport f isheryOther

Subtotal

Upper Mld-Columbla and Snake RiversMcNary DamLower Granite DamPriest Raplds DamSport fishery areasSnake and ClearwaterRlngoldEntlat and WenatcheoMethow

Homlng sitesOther hatcher1 es

SubtotalUnknown

171 0 14 245 141961 47 72 16 18

0 0 0 102 26

104400

4298

1,6710

22000

177

53

5251

081883

13476

5

Total 2,118 821 402 1,016

171254

88

441

0 147 164 119560 140 334 45810 75 24 350 25 13 5

570 357 535 617

L

c̀o0

9011414

2 9 5

a/ Oworshak stock.b/ Skarrmanla stock.c/ wel Is stock.d/ Chelan stock.

129

Dam did not return to the Snake River system but were recovered as strays in

the upper mid-Columnia River (Table 28).

A possible explanation for the anomaly in behavior of the 1979 Tucannon

SFH release may be high water temperatures. A majority of adults from this

lower river stock of Steehead migrated over Bonneville Dam from June to mid-

July 1981. Adults which were imprinted and continued their migration to the

Snake Kiver were confronted with high water temperatures (ranging from 70% to

7 8 - f ) from 17 July to 15 September. Historically, such temperatures result in

a thermal block to migrating steelhead. In most years, such temperatures

occur for a 2- to 3-week period in late August and early September. Fish

generally hold in the cooler Columbia Kiver below the mouth of the Snake Kiver

until water temperatures in the Snake River begin to drop. For late migrating

fish, a short delay is not a problem. However, a delay of over 2 months as

occurred in 1981 may have been sufficient to induce some of these fish to

continue up the cooler upper mid-Columbia River.

In the Wells-Winthrop experiment a stock of non-indigenous steelhead was

transported from the Wells SFH (WDG) (in Wells SFH water) to a production

release site in the lower Methow River and also to the Winthrop NFH (50 miles

further upstream) where the juveniles were held 2 to 4 d for imprinting before

being released. There was no significant difference in the overall recovery

of fish from these two groups (662 and 568, respectively) when the data from

1978 and 1979 are pooled (Table 28). Also, there was no significant

difference in the numbers of adults which migrated over Priest Rapids Dam from

these two release groups, however, signficantly greater numbers of fish from

the lower Methow Kiver release than from the Winthrop River release were taken

in the targeted sport fishery in the Methow River area (p < 0.01, df = 1) (268

130

and 142 fish, respectively). T h i s difference may have been due to the fact

that the Winthrop River release group was exposed to water in the upper Methow

River, and it is possible that a number of adults may have migrated rapidly

upstream and over-wintered above the traditional sport fishery area at the

mouth of the river.

In the Snake River system, adult recoveries frou the 1978 release of

indigenous upriver steelhead at Dworshak NFh were significantly greater than

recoveries from releases of lower river stock fish in the Grande Ronde River

in 1978 (p < 0.01, df - 1). Juveniles from both of these release groups had

about an equal distance to travel and were exposed to the same hazards

inherent in an Outmigration down the Snake River, but over three times as many

adults returned to Lower Granite D a m on the Snake River from the Dworshak NFH

release than from the Grande Ronde 'River release (Table 28). This information

suggests that even though this lower river stock of steelhead will return

upriver to at least as far as Lower Granite Dam on the Snake River, better

returns of adults to upper river areas would be achieved by using indigenous

upriver stocks of steelhead.

Single ixposure Imprint

The single imprint method of trucking steelhead directly from Dworshak

Hatchery for release below Bonneville Dam was partly successful on this

indigenous upriver stock in that it returned about the same number of adults

to the Dworshak NFh homing site as the natural migration release (300 and i78

fish, respectively, Tables 29 and 28). however, homing was impaired on a

portion of these test smolts as indicated by the significantly greater number

of adults (2.8:1} taken in the Zone 6 Indian fishery compared to the number of

adults taken from the natural migration release (p < 0 . 0 1 df - 1). There was

131

some straying of adults above McNary Dam as indicated by the recoveries of 33

stray fish in the upper Mid-Columbia River compared to no straying from the

natural imprint group.

The single imprint method used with non-indigenous upriver stock was

unsuccessful in returning adults to the Winthrop NFh homing site, but did

implant a limited homing cue in a portion of these fish, which in combination

with ‘nigher survival enabled significantly greater numbers of them to return

as adults to the upper Columbia River than fish from the natural migration

(P < 0.01, df q- 1). Between 343 and 374 fish from the single imprint groups

were observed up the river as far as McNary Dam in comparison to only 67 to 8 0

fish from the natural migration groups (Tables 29 and 26). Fish trucked to

ringold returned at a significantly (P < 0.01, df - 1) higher rate over Priest

Rapids D a m than fish from the natural migration groups and also contributed

more fish to the Methowq diver sport fishery than the *Winthrop NFH natural

migration group. However, fish which were trucked to bonneville Dam returned

over Priest 'Rapids Dam at a rate conparable to fish from the natural migration

groups and contributed far fewer fish to the Merthow ‘River sport fishery--

indicating homing was considerably less than the Ringold release.

Many of the juveniles trucked below Bonneville Dam were apparently not

ready or able to accept a honing cue at the time they were transported. These

fish returned as adults to the area near their point of release near

Bonneville Dam and remained there over winter instead of continuing their

migration upriver. As a result, they were vulnerable to both the fall and

winter Indian gill net fishery. In contrast, most of the control releases

were imprinted, migrating rapidly upstream and were much less vulnerable to

the Zone 6 fishery. This was demonstrated by the recovery of 213 transported

132

fish vs 75 natural migration fish from the 1978 Dworshak experiment and 332

trucked vs 59 natural migration fish from the 1973-79 Wells-Winthrop

experiment. Recoveries at Lower Granite and McNary Dams indicated that few if

any of the indigenous and non-indigenous upriver stock steelhead which over

wintered in the Bonneville Dam area continued their upriver migration to the

Snake or upper mid-Columbia Rivers in the spring.

Sequential Exposure Imprint

Two sequential type imprint methods were used with indigenous upriver

stock (Dworshak) steelhead. In 1978, Dworshak NFH stock steelhead were

transported by truck to a barge on the Clearwater giver in Lewiston, Idaho;

held overnight; and then barged through the normal migration route to below

Bonneville Dam. This method was successful in that it returned significantly

more adults to the Dworshak N F H homing site than returned from the natural

migration release (429 and 278 fish, respectively, Tables 28 and 36)

(P < 6.61, df - 1). Even so, homing was impaired for a portion of these

barged fish as evidenced by a T/C ratio of 3.5;1 in the lower river compared

to 1.54:1 for fish returning to the Dworshak NFH homing site. The higher

survival of transported fish as evidenced by 3.5 times as many fish returning

to the lower river as control releases is even more impressive when one

considers that approximately 675 of the control fish surviving to Lower

Granite Dam were provided additional survival enhancement by being transported

around hydroelectric dams on the Snake and Columbia givers.

Another sequential imprint strategy used with Dworshak stock utilized

marked fish that had voluntarily migrated out of ponds, were hauled by truck

to Lower Granite Dam on the Snake River, and then were transferred to barges

or trucks for transport to the lower Columbia River (Bjornn and Binge 1984)

133

(see Dworshak N F H 1980 in this report). This method was not as successful in

returning adults to the Dworshak NFH homing site. Significantly more fish

from the naturally migrating group than from the sequentially imprinted group

returned to the Dworshak NFH homing site (P < 0.01, df - 1). Impaired homing

on a large portion of these transported fish was evidenced by adult recovery

ratios of over 5:1 in the lower river compared to 0.069:1 at the homing site

(Tables 28 and 30).

Our sequential imprint techniques were unsuccessful in returning

non-indigenous stock adults to specific upriver homing sites. However, they

did implant a limited homing cue in a portion of these fish which, in

combination with higher survival, enabled as many or greater numbers of them

to return as adults to areas above McNary Dam as those from the natural

migration groups. Upriver stock adults from the Chelan-Leavenworth tests

returned over McNary Dam ( R M 292) at a rate similar to fish from the natural

migration groups (control) whereas fish from the Wells-Winthrop tests returned

at a significantly greater rate (P < 0.05, df - 1). As previously discussed,

the most apparent loss of homing for these sequentially imprinted (barged)

fish occurred in the 104-mile section of river between McNary and Priest

Rapids Dams. This was verified by the low T/C ratios at Priest Rapids Dam,

the low number of test fish recovered in the sport fishery above the dam, and

the large number of test fish taken in the sport fishery between Ringold and

Priest Rapids Dam. These data suggest that a portion of the fish in these

test lots imprinted to this section of the Columbia River when they were

loaded and held in the barge as juveniles.

134

Straying of adults above McNary Dam to rivers outside the drainage area

of the homing site was minimal. This was indicated by an average of only 4

fish from the Dworshak and Tucannon lots recovered in the upper mid-Columbia

River area and an average of 17 fish from the Winthrop and Leavenworth lots

recovered in the Snake River area.

homing was impaired on portions of the various test groups barged to

below Bonnevile Dam. Their behavior was similar to the single imprint

trucked groups. These adults also returned to the area near their point of

release near Bonneville Dam and remained there over the winter where Many were

captured in the Zone 6 fishery. Yew, if any, of these fish resumed their

migration to the upper river areas.


1. Indigenous upriver stocks of fish will provide a higher rate of

return than non-indigenous stocks back to upriver natcheries.

2. Non-indigenous steelhead given a single exposure imprint and trucked

from the Methow River to below Priest Rapids Dam for release at Singold (with

enhanced survival and partial homing) should return significantly more

steelhead above Priest Rapids Dam than fish released directly into the Methow

River. The net result is a greater contribution of those fish to a sport

fishery all the way from Rngold to the Methwo River.

3. Non-indigenous groups of steelhead given either a single (trucked

directly from the hatchery to below Bonneville Dam) or sequential (trucked to

Richland, Washington, then barged to below Bonneville Dam) exposure imprint

(with enhanced survival and limited homing) will provide significantly more

fish to the sport fishery between Ringold and Priest Rapids Dam than those

released directly into the Methow River. In addition, because of the impaired

135

homing, use of this technique will provide a significant contribution to the

Zone 6 fishery.

4. Sequential exposure imprint techniques used to barge indigenous

stocks of steelhead from Dworshak Hatchery can return over 1.5 times as many

fish back to the hatchery and over twice as many fish to the fisheries as

those released directly from the hatchery.

CONCLUSIONS MD RECOMMENDATIONS

1. Homing cues can be implanted in juvenile salmonids.

2. Imprinting will range from very successful to partial or limited to

uosuccessf ul.

3. Sequential imprinting using truck and/or barge transportation will

likely be more effective than the single imprint method.

4. There are species differences in responding to homing cues.

5. Indigenous stocks are more likely to return to a homing site than

are non-indigenous stocks.

6. Mode of transportation can impact the degree of success for

imprinting fish to return to a specific homing site.

7. Because imprinted fish can be induced to return to different areas,

the fish can be manipulated to contrubute to specific fisheries that are

presently underdeveloped wi tith existing hatchery releases.

a . Even when homing is impaired, strategies can be developed that will

increase net benefits to the fisheries as a whole.

9. Selective imprint strategies plus enhanced survival due to

transportation around high risk areas can provide a tool that Columbia River

fisheries managers can use to provide more salmon and steelhead to various

user groups.

136

10. The research reported here should be used as a baseline for further

research to fine tune the requirements for imprinting specific species and

stocks of salmonids to develop the most effective techniques for reaching

desired mangement goals.

137

The extensive scope of our marking and recovery program was made possible

by the interest and cooperative effort of National Marine Fisheries, the U.S.

Fish and Wildlife Service, Oregon Department of Fish and Wildlife,

Washington's Departments of Wildlife and Fisheries, and the Idaho Department

of Fish and Game in providing both fish and facilities for our experiments and

adult recovery efforts in both the ocean and river fisheries. Additional

ocean recoveries were provided by California Department of Fish and Game,

Alaska Department of Fish and Game, and the Canadian Fisheries Service. The

U.S. Army Corps of Engineers provided the fish barge and facilities at dams on

the Columbia and Snake Rivers. Financial support for this research came from

the region's electrical ratepayers through the Bonneville Power

Administration.

The authors thank M r . Larry G. Brown, WDG, for special assistance and

contribution to study design for steelhead experiments in the mid-Columbia

River region. We also wish to express our appreciation to MR. Carl Utzinger

for his assistance and permission to use the Stavebolt Creek homing site.

138

LITERATURE CITED

Alderice D. F., J. R. Brett, D.R. FIdler, and U. Fagerland.1954 Further observations on olfactory perception in migrating adultcoho and spring salmon-properties of the repellent in mammalian skin.Prog. Repts. Pac. Coast Stations, No. 9 8 P. 10-12

Bishop, Y M. m ., S. E. Fienberg, and P. W. Holland.1975 Discrete multivariate analysis. The Mit Press, Cambridge, Mass.

Bjornn, T. C. and R. K. Ringe.1984. doming of hatchery salmon and steelhead allowed a short-distancevoluntary migration before transport to the lower Columbia River.Final R e p t to NMFS (contract 80-abc-00115), Tech. Rept. 84-i, IdahoCooperative Fishery Research Unit, Univ. of Idaho, 27 p.

Brannon, E. L., R. P. Whitman, and T. P. Quinn.1984. Responses of returning adult coho salmon to home water andpopultion specific odors. Trans. Am. Fish. Soc. li3:37&J77.

Collins, G. B.1976 Effects of dams on Pacific salmon and steelhead trout. Mar. Fish.Rev. 38(1):39-46

Cooper, J. C. and A. T. Scholz.1976. Honing of artificially imprinted steelhead trout. J. Fish. Res.Board Can. 33:826-829

Cooper, J. C., A. T. Scholz, K. M. Horrall, A. D. Hasler, and D. M. Madison.1976. Experimental confirmation of the olfactory hypothesis with

artifically imprinted homing coho salmon, Oncorhynchus kisutch. J.- - - - - - - -Fish. Res. Board Can. 33:703-710

Dawley, E. H., C. W. Sims, R. D. Ledgeruood, D. R. Miller, and f. P. Thrower.1979 A study to define the migrational characteristics of chinook andcoho salmon and steelhead trout in the Columbia River estuary. U.S.Dep. of Commer.,, Natl. Oceanic Atmos. Admin., Nati. Mar. Fish. Serv.,Northwest and Alaska Fish. Cent., Seattle, Wash. 36 p. plus Appendix.Project 712.

Dawley, E. m. , C. W. Sims, r. D. Ledgerwood, D. R. Miller and J. G. W i l l i a m s1981`. A study to define the migrational characteristics of chinook andcoho salmon in the `Columbia River estuary and associated marine water.U.S. Dep. of Commer., Natl. Oceanic Atmos. Admin., Natl. Mar. Fish.Serv., Northwest and Alaska Fish. Cent., Seattle, Wash. 68 p. plusAppendixes. (Report to Pacific Northwest Regional Commission, August1981).

Ebel, W. J.1970. Effect of release location of survival of juvenile fall chinook

salmon, Oncorhynchus Trans. Am. Fish. Soc.:99(4);672-676

tshauytscha_.__-__

139

Ebel, W. J., D. L. Park, and R. C. Johnsen.1973. Effects of transportation on survival and homing of Snake River

chinook salmon and steelhead trout. Fish Bull., U.S. 71(2):549-563.

Ebel, W. J.1974. marking fish and invertebrates. 111. Coded wire tags useful inautomatic recovery of chinook salmon and steelhead trout. Mar. Fish.Rev. 36(7):10-13.

Ebel, W. J.1980. Transportation of chinook salmon, Oncorhynchus tshawytscha, andsteelhead, Salmon gairdneri, smolts in the Colmbia River and effect onadult returns. Fish. Bull., U . S . 78(2):491-505.

Gilbreath, L. G., L. R. Basham, and E. Slatick.1976. Distribution, age, and size of tagged adult steelhead trout inthe Snake River drainage. Mar. Fish. Rev. 39(6):14-H.

Hasler, A. D. and W. J. Wishy.1951. Discrimination of stream odors by fishes and relation to parent

stream behavior. Am. Nat. 85:223-238.

Hisata, J. S., M. L. Schuck, H. W. Wobbs, T. Y. Cho, G. R. Hartinsen,U. Rasmussen, 8. T. Kurose, and W. T. Pederson.

1979-80. Columbia River and tributary tag recovery. Wash. Dep. Game,Olympia, Washington. Project Report sumbitted to the Pacific NorthwstRegional Commission under P.N.R.C. Grant No. 10990647 and NKFS,Contract N. BO-ABC-00039. 291 p. Processed.

Idler, D. R., U. H. M. Fagerlund, and J. Mayoh.1956. Olfactory perception in migrating salmon. I. L-serine, a salmon

repellent in mammalian skin. J. Gen. Physical., 3996):889-892

Idler, D. R., J. R. McBride, R. E. E. Jones, and N. Tomlinson.1961. Olfactory perception in migrating salmon. II. Studies on alaboratory bio-assay for homestream water and mammalian repellent.Can. J. Biochem. Physiol. 39:1575-1584.

Jensen, A. and R. Duncan.1971. Homing in transplanted coho salmon. Prog. Fish. Cult. 33:216-218.

Lister, D. B., D. G. Hickey, and I. Wallace.1981. Review of the effects of enhancement strategies on the homing,straying and survival of Pacific salmonids. Volume I and II. Contract05sb.Fp501-0-1303 Report to Dept. of Fisheries and Ocean, Vancouver,B.C. 51 p. plus Appendixes.

Madison, D. M., A. T. Scholz, J. C. Cooper, and A. D. Hasler.1973. I. Olfactory hypotheses and salmon migration: a synopsis of

recent findings. Fish. Res. Board Can. Tech. Rep. No. 414:37 p.

140

Matthews, G. M., G. A. Swan, and J. R. Smith.1977. Improved bypass and collection system for protection of jwenilesalmon and steelhead trout at Lower Granite Dam. Mar. Fish. Rev.39(7):10-14.

McBride, J. R., D. R. Idler, R. E. E. Jones, and N. Tomlinson.1962. Olfactory perception i n* juvenile salmon. I. Observations on

response of juvenile sockeye to extracts of foods. J. Fish. Res. BoardCan. 19( 2) ̀ :327-334.

McCabe, G. T., Jr., C. W. Long, and D. L. Park.1979. Barge transportation of juvenile salmonids on the Columbia andSnake Rivers, 1977. Mar. Fish Rev., July 1979, 41(7):28-34.

McCabe, G. T., Jr., C. W. Long, and S. L. Leek.1983. Survival and homing of juvenile coho salmon, Oncorhynchus kisutch,transported by barge. Fish. Bull. 81(2):412-415.

Mighell, J. L.1969. Rapid cold branding of salmon and trout with liquid nitrogen.J. Fish. Res. Bd. Can. 26(10):2765-2769.

Nordeng, H.1971. Is the local orientation of anadromous fishes determined bypheromones? Nature, Lond. 233:411-413.

Nordeng, H.1977. A pheromone hypothesis for homeward migration fo anadromoussalmonids. Oikos 28:155-159.

Park, D. L., and W. J. Ebel.1974. Marking fishes and invertebrates. II. Brand size andconfiguration in relation to long term retention on steelhead trout andchinook salmon. Mar. Fish. Rev. 36(7): 7-9.

Park, D. L., J. R. Smith, G. M. Matthews, L. R. Basham, G. A. Swan, G. T.McCabe, Jr., T. E. Ruehle, J. R. Harmon, G. T. McCabe, and B. H. Monk.

1979. Transportation actfvities and related research at Lower Granite,Little Goose, and McNary Dams, 1978. U.S. Dep. of Commer., Natl.Oceanic Amos. admin., Natl. Mar. Fish. Serv., Northwest and AlaskaFish. Cent., Seattle, Wash. 66 p. plus Appendix. (Report to U.S. ArmyCorps of Engineers, Contract DACW68-78-C-0051).

Park, D. L., J. R. Harmon, B. H. Monk, T. E. Ruehle, T. W. Newcomb, L. R.Basham, and T. A. Flagg.

1981. Transportation research on the Columbia and Snake Rivers, 1980.U.S. Dep. of COrnmar., Natl. Oceanic Atmos. Admin., Natl. Mar. Fish.Serv., Northwest and Alaska Fish. Cent., Seattle, Wash. 45 p. plusAppendix. (Report to U.S. Army Corps of Engineers, ContractDAm68-78-C-0051).

141

Park, D. L., J. R. Smith, G. M. Matthews, T. E. Ruehle, J. R. Harmon, S.Achord, and B. H. Monk.

1982. Transportation operations and research on the Snake and ColumbiaRivers, 1981. U.S. Dep. of Commer., Natl. Oceanic Atoms. Admin., Natl.Mar. Fish. Serv., Northwest and Alaska Fish. Cent., Seattle, Wash. 34p. plus Appendixes. (Report to U.S. Army Corps of Engineers, ContractDACW68-78-C-0051).

Raymond, H. L.1979. Effects of dams and impoundments on migrations of juvenile chinooksalmon and steelhead from the Snake River, 1966 to 1975. Trans.American Fisheries Society, 108(6): 505-529.

Raymond, H. L., and G. B. Collins.1974. Techniques for appraisal of migrating juvenile anadromous fishpopulation in the Columbia River Basin. EIFAC/74/1 Symp. 24; January1974.

Raymond, H. L. and C. W. Sims.1980. Assessment of smolt migration and passage enhancement studies for

1979. U.S. Dep. of Commer., Natl. Oceanic Atmos. Admin., Natl. Mar.Fish. Serv., Northwest and Alaska Fish. Cent., Seattle, WA. 48 p. plusAppendix (Report to U.S. Army Corps of Engineers, Contract DACW68-78-C-0051 and DACWS7-079-F-0411).

Scholz, A. T., R. M. Horral, J. C. Cooper, A. D. Hasler, D. M. Madison, R. J.Poff, and R. Daly.

1975. Artificial imprinting of salmon and trout in Lake Michigan. Wis.Dep. Nat. Res. Fish. Manag. Rep. BO:46 pm

Schuck, M. L., M. W. Mobbs, G. V. Lom, T. Y. Cho, R. G. Bfsordi, andJ. W. Ebel.

1980-81. Columbia River and tributary tag recovery. Wash. Dept. Game,Olympia, Washington. Project Report submited to the Pacific NorthwestRegional Commission under P.N.R.C. Grant No. 10090053 and NMFS,Contract NO. 80-ABC-00039. 120 p. Processed.

Sims, c. W . , J. G. Williams, D. A. Faurot, R. C. Johosen, and D. A. Brege.1981. Migrational characteristics of juvenile salmon and steelhead inthe Columbia River basin and related passage research at John DayDam. U.S. Dep. of Commer., Natl. Oceanic Atmos. Admin., Natl. Mar.Fish. Serv., Northwest and Alaska Fish. Cent., Seattle, Wash. 61 p.plus Appendixes. (Report to U.S. Army Corps of Engineers, ContractDACW57-80-F-0394 and DACW68-78-C-0051).

Slatick, E., D. L. Park, and W. J. Ebel.1975. Further studies regarding effects of transportation on survivaland homoing on Snake River chinook salmon and steelhead trout. U.S.Dept. Commer., NOM, NMFS, NWAFC, Fish Bull. 73(4):925-931.

Slatick, E.1975. Laboratory evaluation of a &oil-type steeppass fishway withvarious entrance and exit conditions for passage of adult salmonids andAmerican shad. Mar. Fish. Rev. 37(9):17-26.

142

Slatick, E.1976. Comparative retention of dart and jaw tags on chinook salmon andsteelhead trout tagged on their spawning migration. Mar. Fish.Rev 38(7):24-26.

Slatick, E., A. J. Novotny, and L. G. Gilbreath.1979. Imprinting salmon and steelhead trout for homing. U.S. Dep. of

Commer., Natl. Oceanic Atmos. Admin., Natl. Mar. Fish. Serv., Northwestand Alaska Fish. Cent., Seattle, WA. 2 3 p. (Report to BonnevillePower Administration. Contract 78-l).

Slatick, E., L. G. Gilbreath, and K. A. Walch.1980. Imprinting salmon and steelhead trout for homing, 1979. U.S. Dep.of Commer., Natl. Oceanic Atmos. Admin., Natl. Mar. Fish. Serv.,Northwest and Alaska Fish. Cent., Seattle, WA.(Report

38 pm plus Appendixt0 Bonneville Power Administration, Contract

DE-A179-79-BP-10682).

Sla:ick, E., L. G. `Gilbreath, and K. A. Walch.1981a. Imprinting salmon and steelhead trout for homing, 1980. U.S.Dep. of Commer., Natl. Oceanic Atmos. Admin., Natl. Mar. Fish. Serv.,Northwest and Alaska Fish. Cent., Seattle, WA. 54 p. plus Appendix.(Report BonnevilleDE-A179-SO-BEg236).

Power Administration, Contract

Slatick, E., L. G. Gilbreath, and J. R. Harmon.1981b. Imprinting steelhead for homing. Proceedings of the Salmon andTrout Migratory Behavior Symposium. E. L. Brannoo and E. 0. Salo,editors. First International Symposium, School of Fisheries, Univ. ofWash., Seattle, WA., 1981, p. 247-263.

Slatick, E., L. G. Gilbreath, J. R. Harmon, and K. A. Walsh.1982. Imprinting salmon and steelhead for homing, 1981.

Couuner.,U.S. Dep. of

Natl. Oceanic Atmos. Admin., Natl. Mar. Fish. Serv., Northwestand Alaska Fish. Cent., Seattle, WA.Bonneville Power Administration,

52 p. plus Appendixes (Report toContract DE-A179-BlBP27891).

Slatick, E., L. G. Gilbreath, and J. R. Harmon.1983. Imprinting salmon and steelhead trout for homing, 1982. U.S. Dep.of cp,,er., Natl. Oceanic Atms. Admin., Natl. Mar. Fish. ServNorthwest and Alaska Fish. Cent., Seattle, WA.(Report

46 p= plus Appendix;:Bonneville

DE-A179-glBP:07S91-M001).Power Administration, Contract

Slatick, E., L. G. Gilbreath, J. R. Harmon, C. S. McCutcheon, T. C. Bjornn,and R. R. Rioge.

1984. Imprinting salmon and steelhead trout for homing, 1983. U.S. Dep.of COmmer., Natl. Oceanic Atmos. Admin., Natl. Mar. Fish. Serv.,Northwest and Alaska Fish. Cent., Seattle, WA. 66 p. plus Appendixes(ReportDE-A179+2BP:09646).

Bonneville Power Administration, Contract

143

Smith, J. R. and W. J. Ebel.1973. Aircraft-refueling trailer modified to haul salmon and trout.Mar. Fish. Rev. 35(8):37-40.

Sokal, R. R. and F. Rohlf.1981. Biometry. W. H. Freeman, San Francisco, California.

Taft, A. C. and L.1938. Homingand silver23: i 18-125.

Shapovalov.instinct and straying among steelhead, Salmo gairdnerii,salmon, Dncorhynchus kisutch. Calif. Fish and Game,

Vreeland, R. R., R. J. Wahle, and A. H. Arp.- 1975. Homing behavior and contribution to Columbia River fisheries of

marked coho salmon released at two locations. Fish. Bull. 73:717-725.

Wagner, H. H.1968. Effect of stocking time of survival of steelhead trout, Salmogairdnerii, in Oregon. Trans. Am. Fish. Sot. 97(4):374-379.

Wahle, R. J. and E. ahaney.1981. Establishment of noofodigeoous runs of spring chinook salmon,Dncorhynchus tshawytscha, fn -the Wind River drainage of the ColumbiaRiver, 1955-63. U.S. Fish. Bull. 79(3):507-516.

Zaugg, W. S. and L. R. McLain.1970. Adeoosinetriphosphatase activity in gills of salmoofds: seasonalvariation and salt water influence in coho salmon, Olncorhyochuskisutch. comp. Biochem. Physiol. 35, p. 587-596.

Zaugg, W. S., B. L. Adams, and L. R. McLain.1972. Steelhead migration: potential temperature effects as indicatedby gill adeoosine triphosphatase activities. Science. 176,p. 415-416.

Zimmer, P. D., R. J. Wahle, and E. M. Maltzeff.1963. Progress report spring chinook salmon transplantation study. U.S.Fish. Wildl. Serv., Spec. Sci. Rept. Fish. 443. 2 4 p.

imprinting hatchery reared salmon and steelhead...

Documents